update redis data

README.md

@@ -0,0 +1,3 @@
+Asi debe estar en el .env para el agente de web search
+
+TAVILY_API_KEY=

ai-elements-tools.md

@@ -0,0 +1,384 @@
+# Generative User Interfaces
+
+Generative user interfaces (generative UI) is the process of allowing a large language model (LLM) to go beyond text and "generate UI". This creates a more engaging and AI-native experience for users.
+
+<WeatherSearch />
+
+At the core of generative UI are [ tools ](/docs/ai-sdk-core/tools-and-tool-calling), which are functions you provide to the model to perform specialized tasks like getting the weather in a location. The model can decide when and how to use these tools based on the context of the conversation.
+
+Generative UI is the process of connecting the results of a tool call to a React component. Here's how it works:
+
+1. You provide the model with a prompt or conversation history, along with a set of tools.
+2. Based on the context, the model may decide to call a tool.
+3. If a tool is called, it will execute and return data.
+4. This data can then be passed to a React component for rendering.
+
+By passing the tool results to React components, you can create a generative UI experience that's more engaging and adaptive to your needs.
+
+## Build a Generative UI Chat Interface
+
+Let's create a chat interface that handles text-based conversations and incorporates dynamic UI elements based on model responses.
+
+### Basic Chat Implementation
+
+Start with a basic chat implementation using the `useChat` hook:
+
+```tsx filename="app/page.tsx"
+'use client';
+
+import { useChat } from '@ai-sdk/react';
+import { useState } from 'react';
+
+export default function Page() {
+  const [input, setInput] = useState('');
+  const { messages, sendMessage } = useChat();
+
+  const handleSubmit = (e: React.FormEvent) => {
+    e.preventDefault();
+    sendMessage({ text: input });
+    setInput('');
+  };
+
+  return (
+    <div>
+      {messages.map(message => (
+        <div key={message.id}>
+          <div>{message.role === 'user' ? 'User: ' : 'AI: '}</div>
+          <div>
+            {message.parts.map((part, index) => {
+              if (part.type === 'text') {
+                return <span key={index}>{part.text}</span>;
+              }
+              return null;
+            })}
+          </div>
+        </div>
+      ))}
+
+      <form onSubmit={handleSubmit}>
+        <input
+          value={input}
+          onChange={e => setInput(e.target.value)}
+          placeholder="Type a message..."
+        />
+        <button type="submit">Send</button>
+      </form>
+    </div>
+  );
+}
+```
+
+To handle the chat requests and model responses, set up an API route:
+
+```ts filename="app/api/chat/route.ts"
+import { openai } from '@ai-sdk/openai';
+import { streamText, convertToModelMessages, UIMessage, stepCountIs } from 'ai';
+
+export async function POST(request: Request) {
+  const { messages }: { messages: UIMessage[] } = await request.json();
+
+  const result = streamText({
+    model: openai('gpt-4o'),
+    system: 'You are a friendly assistant!',
+    messages: convertToModelMessages(messages),
+    stopWhen: stepCountIs(5),
+  });
+
+  return result.toUIMessageStreamResponse();
+}
+```
+
+This API route uses the `streamText` function to process chat messages and stream the model's responses back to the client.
+
+### Create a Tool
+
+Before enhancing your chat interface with dynamic UI elements, you need to create a tool and corresponding React component. A tool will allow the model to perform a specific action, such as fetching weather information.
+
+Create a new file called `ai/tools.ts` with the following content:
+
+```ts filename="ai/tools.ts"
+import { tool as createTool } from 'ai';
+import { z } from 'zod';
+
+export const weatherTool = createTool({
+  description: 'Display the weather for a location',
+  inputSchema: z.object({
+    location: z.string().describe('The location to get the weather for'),
+  }),
+  execute: async function ({ location }) {
+    await new Promise(resolve => setTimeout(resolve, 2000));
+    return { weather: 'Sunny', temperature: 75, location };
+  },
+});
+
+export const tools = {
+  displayWeather: weatherTool,
+};
+```
+
+In this file, you've created a tool called `weatherTool`. This tool simulates fetching weather information for a given location. This tool will return simulated data after a 2-second delay. In a real-world application, you would replace this simulation with an actual API call to a weather service.
+
+### Update the API Route
+
+Update the API route to include the tool you've defined:
+
+```ts filename="app/api/chat/route.ts" highlight="3,8,14"
+import { openai } from '@ai-sdk/openai';
+import { streamText, convertToModelMessages, UIMessage, stepCountIs } from 'ai';
+import { tools } from '@/ai/tools';
+
+export async function POST(request: Request) {
+  const { messages }: { messages: UIMessage[] } = await request.json();
+
+  const result = streamText({
+    model: openai('gpt-4o'),
+    system: 'You are a friendly assistant!',
+    messages: convertToModelMessages(messages),
+    stopWhen: stepCountIs(5),
+    tools,
+  });
+
+  return result.toUIMessageStreamResponse();
+}
+```
+
+Now that you've defined the tool and added it to your `streamText` call, let's build a React component to display the weather information it returns.
+
+### Create UI Components
+
+Create a new file called `components/weather.tsx`:
+
+```tsx filename="components/weather.tsx"
+type WeatherProps = {
+  temperature: number;
+  weather: string;
+  location: string;
+};
+
+export const Weather = ({ temperature, weather, location }: WeatherProps) => {
+  return (
+    <div>
+      <h2>Current Weather for {location}</h2>
+      <p>Condition: {weather}</p>
+      <p>Temperature: {temperature}°C</p>
+    </div>
+  );
+};
+```
+
+This component will display the weather information for a given location. It takes three props: `temperature`, `weather`, and `location` (exactly what the `weatherTool` returns).
+
+### Render the Weather Component
+
+Now that you have your tool and corresponding React component, let's integrate them into your chat interface. You'll render the Weather component when the model calls the weather tool.
+
+To check if the model has called a tool, you can check the `parts` array of the UIMessage object for tool-specific parts. In AI SDK 5.0, tool parts use typed naming: `tool-${toolName}` instead of generic types.
+
+Update your `page.tsx` file:
+
+```tsx filename="app/page.tsx" highlight="4,9,14-15,19-46"
+'use client';
+
+import { useChat } from '@ai-sdk/react';
+import { useState } from 'react';
+import { Weather } from '@/components/weather';
+
+export default function Page() {
+  const [input, setInput] = useState('');
+  const { messages, sendMessage } = useChat();
+
+  const handleSubmit = (e: React.FormEvent) => {
+    e.preventDefault();
+    sendMessage({ text: input });
+    setInput('');
+  };
+
+  return (
+    <div>
+      {messages.map(message => (
+        <div key={message.id}>
+          <div>{message.role === 'user' ? 'User: ' : 'AI: '}</div>
+          <div>
+            {message.parts.map((part, index) => {
+              if (part.type === 'text') {
+                return <span key={index}>{part.text}</span>;
+              }
+
+              if (part.type === 'tool-displayWeather') {
+                switch (part.state) {
+                  case 'input-available':
+                    return <div key={index}>Loading weather...</div>;
+                  case 'output-available':
+                    return (
+                      <div key={index}>
+                        <Weather {...part.output} />
+                      </div>
+                    );
+                  case 'output-error':
+                    return <div key={index}>Error: {part.errorText}</div>;
+                  default:
+                    return null;
+                }
+              }
+
+              return null;
+            })}
+          </div>
+        </div>
+      ))}
+
+      <form onSubmit={handleSubmit}>
+        <input
+          value={input}
+          onChange={e => setInput(e.target.value)}
+          placeholder="Type a message..."
+        />
+        <button type="submit">Send</button>
+      </form>
+    </div>
+  );
+}
+```
+
+In this updated code snippet, you:
+
+1. Use manual input state management with `useState` instead of the built-in `input` and `handleInputChange`.
+2. Use `sendMessage` instead of `handleSubmit` to send messages.
+3. Check the `parts` array of each message for different content types.
+4. Handle tool parts with type `tool-displayWeather` and their different states (`input-available`, `output-available`, `output-error`).
+
+This approach allows you to dynamically render UI components based on the model's responses, creating a more interactive and context-aware chat experience.
+
+## Expanding Your Generative UI Application
+
+You can enhance your chat application by adding more tools and components, creating a richer and more versatile user experience. Here's how you can expand your application:
+
+### Adding More Tools
+
+To add more tools, simply define them in your `ai/tools.ts` file:
+
+```ts
+// Add a new stock tool
+export const stockTool = createTool({
+  description: 'Get price for a stock',
+  inputSchema: z.object({
+    symbol: z.string().describe('The stock symbol to get the price for'),
+  }),
+  execute: async function ({ symbol }) {
+    // Simulated API call
+    await new Promise(resolve => setTimeout(resolve, 2000));
+    return { symbol, price: 100 };
+  },
+});
+
+// Update the tools object
+export const tools = {
+  displayWeather: weatherTool,
+  getStockPrice: stockTool,
+};
+```
+
+Now, create a new file called `components/stock.tsx`:
+
+```tsx
+type StockProps = {
+  price: number;
+  symbol: string;
+};
+
+export const Stock = ({ price, symbol }: StockProps) => {
+  return (
+    <div>
+      <h2>Stock Information</h2>
+      <p>Symbol: {symbol}</p>
+      <p>Price: ${price}</p>
+    </div>
+  );
+};
+```
+
+Finally, update your `page.tsx` file to include the new Stock component:
+
+```tsx
+'use client';
+
+import { useChat } from '@ai-sdk/react';
+import { useState } from 'react';
+import { Weather } from '@/components/weather';
+import { Stock } from '@/components/stock';
+
+export default function Page() {
+  const [input, setInput] = useState('');
+  const { messages, sendMessage } = useChat();
+
+  const handleSubmit = (e: React.FormEvent) => {
+    e.preventDefault();
+    sendMessage({ text: input });
+    setInput('');
+  };
+
+  return (
+    <div>
+      {messages.map(message => (
+        <div key={message.id}>
+          <div>{message.role}</div>
+          <div>
+            {message.parts.map((part, index) => {
+              if (part.type === 'text') {
+                return <span key={index}>{part.text}</span>;
+              }
+
+              if (part.type === 'tool-displayWeather') {
+                switch (part.state) {
+                  case 'input-available':
+                    return <div key={index}>Loading weather...</div>;
+                  case 'output-available':
+                    return (
+                      <div key={index}>
+                        <Weather {...part.output} />
+                      </div>
+                    );
+                  case 'output-error':
+                    return <div key={index}>Error: {part.errorText}</div>;
+                  default:
+                    return null;
+                }
+              }
+
+              if (part.type === 'tool-getStockPrice') {
+                switch (part.state) {
+                  case 'input-available':
+                    return <div key={index}>Loading stock price...</div>;
+                  case 'output-available':
+                    return (
+                      <div key={index}>
+                        <Stock {...part.output} />
+                      </div>
+                    );
+                  case 'output-error':
+                    return <div key={index}>Error: {part.errorText}</div>;
+                  default:
+                    return null;
+                }
+              }
+
+              return null;
+            })}
+          </div>
+        </div>
+      ))}
+
+      <form onSubmit={handleSubmit}>
+        <input
+          type="text"
+          value={input}
+          onChange={e => setInput(e.target.value)}
+        />
+        <button type="submit">Send</button>
+      </form>
+    </div>
+  );
+}
+```
+
+By following this pattern, you can continue to add more tools and components, expanding the capabilities of your Generative UI application.

backend/RATE_LIMITING.md

@@ -0,0 +1,105 @@
+# Configuración de Rate Limiting para Azure OpenAI
+
+Este documento explica cómo configurar el rate limiting para evitar errores `429 RateLimitReached` en Azure OpenAI.
+
+## Variables de Entorno
+
+Agrega estas variables en tu archivo `.env`:
+
+```bash
+# Rate limiting para embeddings
+EMBEDDING_BATCH_SIZE=16
+EMBEDDING_DELAY_BETWEEN_BATCHES=1.0
+EMBEDDING_MAX_RETRIES=5
+```
+
+## Configuración según Azure OpenAI Tier
+
+### **S0 Tier (Gratis/Básico)**
+- **Límite**: ~1-3 requests/minuto, ~1,000 tokens/minuto
+- **Configuración recomendada**:
+  ```bash
+  EMBEDDING_BATCH_SIZE=16
+  EMBEDDING_DELAY_BETWEEN_BATCHES=1.0
+  EMBEDDING_MAX_RETRIES=5
+  ```
+
+### **Standard Tier**
+- **Límite**: ~10-20 requests/segundo, ~100,000 tokens/minuto
+- **Configuración recomendada**:
+  ```bash
+  EMBEDDING_BATCH_SIZE=50
+  EMBEDDING_DELAY_BETWEEN_BATCHES=0.5
+  EMBEDDING_MAX_RETRIES=3
+  ```
+
+### **Premium Tier**
+- **Límite**: ~100+ requests/segundo, ~500,000+ tokens/minuto
+- **Configuración recomendada**:
+  ```bash
+  EMBEDDING_BATCH_SIZE=100
+  EMBEDDING_DELAY_BETWEEN_BATCHES=0.1
+  EMBEDDING_MAX_RETRIES=3
+  ```
+
+## Cómo Funciona el Rate Limiting
+
+### 1. **Batching**
+Los textos se dividen en lotes de tamaño `EMBEDDING_BATCH_SIZE`. Un lote más pequeño reduce la probabilidad de exceder el rate limit.
+
+### 2. **Delays entre Batches**
+Después de procesar cada lote, el sistema espera `EMBEDDING_DELAY_BETWEEN_BATCHES` segundos antes de procesar el siguiente.
+
+### 3. **Retry con Exponential Backoff**
+Si ocurre un error 429 (rate limit):
+- **Reintento 1**: espera 2 segundos
+- **Reintento 2**: espera 4 segundos
+- **Reintento 3**: espera 8 segundos
+- **Reintento 4**: espera 16 segundos
+- **Reintento 5**: espera 32 segundos
+
+Después de `EMBEDDING_MAX_RETRIES` reintentos, el proceso falla.
+
+## Monitoreo de Logs
+
+Cuando procesas documentos, verás logs como:
+
+```
+📊 Procesando batch 1/10 (16 textos)...
+✓ Batch 1/10 completado exitosamente
+📊 Procesando batch 2/10 (16 textos)...
+⚠️  Rate limit alcanzado en batch 2/10. Reintento 1/5 en 2s...
+✓ Batch 2/10 completado exitosamente
+...
+✅ Embeddings generados exitosamente: 150 vectores de 3072D
+```
+
+## Cálculo de Tiempo de Procesamiento
+
+Para estimar cuánto tardará el procesamiento:
+
+```
+Tiempo estimado = (total_chunks / EMBEDDING_BATCH_SIZE) * EMBEDDING_DELAY_BETWEEN_BATCHES
+```
+
+**Ejemplos**:
+- 100 chunks con S0 config: `(100/16) * 1.0 = ~6.25 segundos` (sin contar reintentos)
+- 1000 chunks con S0 config: `(1000/16) * 1.0 = ~62.5 segundos` (sin contar reintentos)
+
+## Ajuste Dinámico
+
+Si experimentas muchos errores 429:
+1. **Reduce** `EMBEDDING_BATCH_SIZE` (ej: de 16 a 8)
+2. **Aumenta** `EMBEDDING_DELAY_BETWEEN_BATCHES` (ej: de 1.0 a 2.0)
+3. **Aumenta** `EMBEDDING_MAX_RETRIES` (ej: de 5 a 10)
+
+Si el procesamiento es muy lento y NO tienes errores 429:
+1. **Aumenta** `EMBEDDING_BATCH_SIZE` (ej: de 16 a 32)
+2. **Reduce** `EMBEDDING_DELAY_BETWEEN_BATCHES` (ej: de 1.0 a 0.5)
+
+## Upgrade de Azure OpenAI Tier
+
+Para aumentar tu límite, visita:
+https://aka.ms/oai/quotaincrease
+
+Después del upgrade, ajusta las variables de entorno según tu nuevo tier.

backend/app/agents/analyst/__init__.py

@@ -0,0 +1,112 @@
+from __future__ import annotations
+
+from typing import Any, Iterable, List
+
+from app.agents.form_auditor.models import ExtractedIrsForm990PfDataSchema
+
+from .agent import agent
+from .metrics import SnapshotBundle, build_key_metrics, build_snapshots
+from .models import AnalystReport, AnalystState
+
+__all__ = ["build_performance_report"]
+
+
+def _resolve_year(
+    entry: dict[str, Any], extraction: ExtractedIrsForm990PfDataSchema
+) -> int:
+    candidates: Iterable[Any] = (
+        entry.get("calendar_year"),
+        entry.get("year"),
+        entry.get("tax_year"),
+        entry.get("return_year"),
+        entry.get("metadata", {}).get("return_year")
+        if isinstance(entry.get("metadata"), dict)
+        else None,
+        entry.get("metadata", {}).get("tax_year")
+        if isinstance(entry.get("metadata"), dict)
+        else None,
+        extraction.core_organization_metadata.calendar_year,
+    )
+    for candidate in candidates:
+        if candidate in (None, ""):
+            continue
+        try:
+            return int(candidate)
+        except (TypeError, ValueError):
+            continue
+    raise ValueError("Unable to determine filing year for one of the payload entries.")
+
+
+async def build_performance_report(payloads: List[dict[str, Any]]) -> AnalystReport:
+    if not payloads:
+        raise ValueError("At least one payload is required for performance analysis.")
+
+    bundles: List[SnapshotBundle] = []
+
+    organisation_name = ""
+    organisation_ein = ""
+
+    for entry in payloads:
+        if not isinstance(entry, dict):
+            raise TypeError("Each payload entry must be a dict.")
+
+        extraction_payload = entry.get("extraction") if "extraction" in entry else entry
+        extraction = ExtractedIrsForm990PfDataSchema.model_validate(extraction_payload)
+        year = _resolve_year(entry, extraction)
+
+        if not organisation_ein:
+            organisation_ein = extraction.core_organization_metadata.ein
+            organisation_name = extraction.core_organization_metadata.legal_name
+        else:
+            if extraction.core_organization_metadata.ein != organisation_ein:
+                raise ValueError(
+                    "All payload entries must belong to the same organization."
+                )
+
+        bundles.append(SnapshotBundle(year=year, extraction=extraction))
+
+    bundles.sort(key=lambda bundle: bundle.year)
+    snapshots = build_snapshots(bundles)
+    metrics = build_key_metrics(snapshots)
+
+    notes = []
+    if metrics:
+        revenue_metric = metrics[0]
+        expense_metric = metrics[1] if len(metrics) > 1 else None
+        if revenue_metric.cagr is not None:
+            notes.append(f"Revenue CAGR: {revenue_metric.cagr:.2%}")
+        if expense_metric and expense_metric.cagr is not None:
+            notes.append(f"Expense CAGR: {expense_metric.cagr:.2%}")
+        surplus_metric = next(
+            (m for m in metrics if m.name == "Operating Surplus"), None
+        )
+        if surplus_metric:
+            last_value = surplus_metric.points[-1].value if surplus_metric.points else 0
+            notes.append(f"Latest operating surplus: {last_value:,.0f}")
+
+    state = AnalystState(
+        organisation_name=organisation_name,
+        organisation_ein=organisation_ein,
+        series=snapshots,
+        key_metrics=metrics,
+        notes=notes,
+    )
+
+    prompt = (
+        "Analyze the provided multi-year financial context. Quantify notable trends, "
+        "call out risks or strengths, and supply actionable recommendations. "
+        "Capture both positive momentum and areas requiring attention."
+    )
+    result = await agent.run(prompt, deps=state)
+    report = result.output
+
+    years = [snapshot.year for snapshot in snapshots]
+
+    return report.model_copy(
+        update={
+            "organisation_name": organisation_name,
+            "organisation_ein": organisation_ein,
+            "years_analyzed": years,
+            "key_metrics": metrics,
+        }
+    )

backend/app/agents/analyst/agent.py

@@ -0,0 +1,33 @@
+from __future__ import annotations
+
+from pydantic_ai import Agent
+from pydantic_ai.models.openai import OpenAIChatModel
+from pydantic_ai.providers.azure import AzureProvider
+
+from app.core.config import settings
+
+from .models import AnalystReport, AnalystState
+
+provider = AzureProvider(
+    azure_endpoint=settings.AZURE_OPENAI_ENDPOINT,
+    api_version=settings.AZURE_OPENAI_API_VERSION,
+    api_key=settings.AZURE_OPENAI_API_KEY,
+)
+
+model = OpenAIChatModel(model_name="gpt-4o", provider=provider)
+
+agent = Agent(
+    model=model,
+    name="MultiYearAnalyst",
+    deps_type=AnalystState,
+    output_type=AnalystReport,
+    system_prompt=(
+        "You are a nonprofit financial analyst. You receive multi-year Form 990 extractions "
+        "summarized into deterministic metrics (series, ratios, surplus, CAGR). Use the context "
+        "to highlight performance trends, governance implications, and forward-looking risks. "
+        "Focus on numeric trends: revenue growth, expense discipline, surplus stability, "
+        "program-vs-admin mix, and fundraising efficiency. Provide concise bullet insights, "
+        "clear recommendations tied to the data, and a balanced outlook (strengths vs watch items). "
+        "Only cite facts available in the provided series—do not invent figures."
+    ),
+)

backend/app/agents/analyst/metrics.py

@@ -0,0 +1,197 @@
+from __future__ import annotations
+
+from dataclasses import dataclass
+from typing import Iterable, List, Sequence, Tuple
+
+from app.agents.form_auditor.models import ExtractedIrsForm990PfDataSchema
+
+from .models import TrendDirection, TrendMetric, TrendMetricPoint, YearlySnapshot
+
+
+@dataclass
+class SnapshotBundle:
+    year: int
+    extraction: ExtractedIrsForm990PfDataSchema
+
+
+def _safe_ratio(numerator: float, denominator: float) -> float | None:
+    if denominator in (0, None):
+        return None
+    try:
+        return numerator / denominator
+    except ZeroDivisionError:
+        return None
+
+
+def _growth(current: float, previous: float | None) -> float | None:
+    if previous in (None, 0):
+        return None
+    try:
+        return (current - previous) / previous
+    except ZeroDivisionError:
+        return None
+
+
+def _direction_from_points(values: Sequence[float | None]) -> TrendDirection:
+    clean = [value for value in values if value is not None]
+    if len(clean) < 2:
+        return TrendDirection.STABLE
+
+    start, end = clean[0], clean[-1]
+    if start is None or end is None:
+        return TrendDirection.STABLE
+
+    delta = end - start
+    tolerance = abs(start) * 0.02 if start else 0.01
+    if abs(delta) <= tolerance:
+        return TrendDirection.STABLE
+
+    if len(clean) > 2:
+        swings = sum(
+            1
+            for idx in range(1, len(clean) - 1)
+            if (clean[idx] - clean[idx - 1]) * (clean[idx + 1] - clean[idx]) < 0
+        )
+        if swings >= len(clean) // 2:
+            return TrendDirection.VOLATILE
+
+    return TrendDirection.IMPROVING if delta > 0 else TrendDirection.DECLINING
+
+
+def _cagr(start: float | None, end: float | None, periods: int) -> float | None:
+    if start is None or end is None or start <= 0 or end <= 0 or periods <= 0:
+        return None
+    return (end / start) ** (1 / periods) - 1
+
+
+def build_snapshots(bundles: Sequence[SnapshotBundle]) -> List[YearlySnapshot]:
+    snapshots: List[YearlySnapshot] = []
+    previous_revenue = None
+    previous_expenses = None
+
+    for bundle in bundles:
+        rev = bundle.extraction.revenue_breakdown.total_revenue
+        exp = bundle.extraction.expenses_breakdown.total_expenses
+        program = bundle.extraction.expenses_breakdown.program_services_expenses
+        admin = bundle.extraction.expenses_breakdown.management_general_expenses
+        fundraising = bundle.extraction.expenses_breakdown.fundraising_expenses
+
+        snapshots.append(
+            YearlySnapshot(
+                year=bundle.year,
+                total_revenue=rev,
+                total_expenses=exp,
+                revenue_growth=_growth(rev, previous_revenue),
+                expense_growth=_growth(exp, previous_expenses),
+                surplus=rev - exp,
+                program_ratio=_safe_ratio(program, exp),
+                admin_ratio=_safe_ratio(admin, exp),
+                fundraising_ratio=_safe_ratio(fundraising, exp),
+                net_margin=_safe_ratio(rev - exp, rev),
+            )
+        )
+        previous_revenue = rev
+        previous_expenses = exp
+
+    return snapshots
+
+
+def _metric_from_series(
+    name: str,
+    unit: str,
+    description: str,
+    values: Iterable[Tuple[int, float | None]],
+) -> TrendMetric:
+    points = [
+        TrendMetricPoint(year=year, value=value or 0.0, growth=None)
+        for year, value in values
+    ]
+
+    for idx in range(1, len(points)):
+        prev = points[idx - 1].value
+        curr = points[idx].value
+        points[idx].growth = _growth(curr, prev)
+
+    data_values = [point.value for point in points]
+    direction = _direction_from_points(data_values)
+    cagr = None
+    if len(points) >= 2:
+        cagr = _cagr(points[0].value, points[-1].value, len(points) - 1)
+
+    return TrendMetric(
+        name=name,
+        unit=unit,
+        description=description,
+        points=points,
+        cagr=cagr,
+        direction=direction,
+    )
+
+
+def build_key_metrics(snapshots: Sequence[YearlySnapshot]) -> List[TrendMetric]:
+    if not snapshots:
+        return []
+
+    metrics = [
+        _metric_from_series(
+            "Total Revenue",
+            "USD",
+            "Reported total revenue in Part I.",
+            [(snap.year, snap.total_revenue) for snap in snapshots],
+        ),
+        _metric_from_series(
+            "Total Expenses",
+            "USD",
+            "Reported total expenses in Part I.",
+            [(snap.year, snap.total_expenses) for snap in snapshots],
+        ),
+        _metric_from_series(
+            "Operating Surplus",
+            "USD",
+            "Difference between total revenue and total expenses.",
+            [(snap.year, snap.surplus) for snap in snapshots],
+        ),
+        _metric_from_series(
+            "Program Service Ratio",
+            "Ratio",
+            "Program service expenses divided by total expenses.",
+            [
+                (
+                    snap.year,
+                    snap.program_ratio if snap.program_ratio is not None else 0.0,
+                )
+                for snap in snapshots
+            ],
+        ),
+        _metric_from_series(
+            "Administrative Ratio",
+            "Ratio",
+            "Management & general expenses divided by total expenses.",
+            [
+                (snap.year, snap.admin_ratio if snap.admin_ratio is not None else 0.0)
+                for snap in snapshots
+            ],
+        ),
+        _metric_from_series(
+            "Fundraising Ratio",
+            "Ratio",
+            "Fundraising expenses divided by total expenses.",
+            [
+                (
+                    snap.year,
+                    snap.fundraising_ratio
+                    if snap.fundraising_ratio is not None
+                    else 0.0,
+                )
+                for snap in snapshots
+            ],
+        ),
+    ]
+
+    for metric in metrics:
+        if metric.name.endswith("Ratio"):
+            metric.notes = "Higher values indicate greater spending share."
+        elif metric.name == "Operating Surplus":
+            metric.notes = "Positive surplus implies revenues exceeded expenses."
+
+    return metrics

backend/app/agents/analyst/models.py

@@ -0,0 +1,74 @@
+from __future__ import annotations
+
+from enum import Enum
+from typing import List
+
+from pydantic import BaseModel, Field
+
+
+class TrendDirection(str, Enum):
+    IMPROVING = "Improving"
+    DECLINING = "Declining"
+    STABLE = "Stable"
+    VOLATILE = "Volatile"
+
+
+class TrendMetricPoint(BaseModel):
+    year: int
+    value: float
+    growth: float | None = Field(
+        default=None, description="Year-over-year growth expressed as a decimal."
+    )
+
+
+class TrendMetric(BaseModel):
+    name: str
+    unit: str
+    description: str
+    points: List[TrendMetricPoint]
+    cagr: float | None = Field(
+        default=None,
+        description="Compound annual growth rate across the analyzed period.",
+    )
+    direction: TrendDirection = Field(
+        default=TrendDirection.STABLE, description="Overall direction of the metric."
+    )
+    notes: str | None = None
+
+
+class TrendInsight(BaseModel):
+    category: str
+    direction: TrendDirection
+    summary: str
+    confidence: float = Field(default=0.7, ge=0.0, le=1.0)
+
+
+class AnalystReport(BaseModel):
+    organisation_name: str
+    organisation_ein: str
+    years_analyzed: List[int] = Field(default_factory=list)
+    key_metrics: List[TrendMetric] = Field(default_factory=list)
+    insights: List[TrendInsight] = Field(default_factory=list)
+    recommendations: List[str] = Field(default_factory=list)
+    outlook: str = "Pending analysis"
+
+
+class YearlySnapshot(BaseModel):
+    year: int
+    total_revenue: float
+    total_expenses: float
+    revenue_growth: float | None = None
+    expense_growth: float | None = None
+    surplus: float | None = None
+    program_ratio: float | None = None
+    admin_ratio: float | None = None
+    fundraising_ratio: float | None = None
+    net_margin: float | None = None
+
+
+class AnalystState(BaseModel):
+    organisation_name: str
+    organisation_ein: str
+    series: List[YearlySnapshot]
+    key_metrics: List[TrendMetric]
+    notes: List[str] = Field(default_factory=list)

backend/app/agents/form_auditor/__init__.py

@@ -0,0 +1,47 @@
+from __future__ import annotations
+
+from typing import Any
+
+from .agent import agent, prepare_initial_findings
+from .models import (
+    AuditReport,
+    ExtractedIrsForm990PfDataSchema,
+    ValidatorState,
+)
+
+
+async def build_audit_report(payload: dict[str, Any]) -> AuditReport:
+    metadata_raw: Any = None
+    extraction_payload: Any = None
+
+    if isinstance(payload, dict) and "extraction" in payload:
+        extraction_payload = payload.get("extraction")
+        metadata_raw = payload.get("metadata")
+    else:
+        extraction_payload = payload
+
+    if extraction_payload is None:
+        raise ValueError("Payload missing extraction data.")
+
+    extraction = ExtractedIrsForm990PfDataSchema.model_validate(extraction_payload)
+
+    initial_findings = prepare_initial_findings(extraction)
+
+    metadata: dict[str, Any] = {}
+    if isinstance(metadata_raw, dict):
+        metadata = {str(k): v for k, v in metadata_raw.items()}
+
+    state = ValidatorState(
+        extraction=extraction,
+        initial_findings=initial_findings,
+        metadata=metadata,
+    )
+
+    prompt = (
+        "Review the Form 990 extraction and deterministic checks. Validate or adjust "
+        "the findings, add any additional issues or mitigations, and craft narrative "
+        "section summaries that highlight the most material points. Focus on concrete "
+        "evidence; do not fabricate figures."
+    )
+    result = await agent.run(prompt, deps=state)
+    return result.output

backend/app/agents/form_auditor/agent.py

@@ -0,0 +1,155 @@
+from __future__ import annotations
+
+from collections.abc import Iterable
+
+from pydantic_ai import Agent, RunContext
+from pydantic_ai.models.openai import OpenAIChatModel
+from pydantic_ai.providers.azure import AzureProvider
+
+from app.core.config import settings
+
+from .checks import (
+    aggregate_findings,
+    build_section_summaries,
+    check_balance_sheet_presence,
+    check_board_engagement,
+    check_expense_totals,
+    check_fundraising_alignment,
+    check_governance_policies,
+    check_missing_operational_details,
+    check_revenue_totals,
+    compose_overall_summary,
+    irs_ein_lookup,
+)
+from .models import (
+    AuditFinding,
+    AuditReport,
+    ExtractedIrsForm990PfDataSchema,
+    Severity,
+    ValidatorState,
+)
+
+provider = AzureProvider(
+    azure_endpoint=settings.AZURE_OPENAI_ENDPOINT,
+    api_version=settings.AZURE_OPENAI_API_VERSION,
+    api_key=settings.AZURE_OPENAI_API_KEY,
+)
+model = OpenAIChatModel(model_name="gpt-4o", provider=provider)
+agent = Agent(model=model)
+
+
+def prepare_initial_findings(
+    extraction: ExtractedIrsForm990PfDataSchema,
+) -> list[AuditFinding]:
+    findings = [
+        check_revenue_totals(extraction),
+        check_expense_totals(extraction),
+        check_fundraising_alignment(extraction),
+        check_balance_sheet_presence(extraction),
+        check_board_engagement(extraction),
+        check_missing_operational_details(extraction),
+    ]
+    findings.extend(check_governance_policies(extraction))
+    return findings
+
+
+def _merge_findings(
+    findings: Iterable[AuditFinding],
+    added: Iterable[AuditFinding],
+) -> list[AuditFinding]:
+    existing = {finding.check_id: finding for finding in findings}
+    for finding in added:
+        existing[finding.check_id] = finding
+    return list(existing.values())
+
+
+agent = Agent(
+    model=model,
+    name="FormValidator",
+    deps_type=ValidatorState,
+    output_type=AuditReport,
+    system_prompt=(
+        "You are a Form 990 auditor. Review the extraction data and deterministic "
+        "checks provided in deps. Use tools to confirm calculations, add or adjust "
+        "findings, supply mitigation guidance, and craft concise section summaries. "
+        "The AuditReport must include severity (`Pass`, `Warning`, `Error`), "
+        "confidence scores, mitigation advice, section summaries, and an overall summary. "
+        "Ground every statement in supplied data; do not invent financial figures."
+    ),
+)
+
+
+@agent.tool
+def revenue_check(ctx: RunContext[ValidatorState]) -> AuditFinding:
+    return check_revenue_totals(ctx.deps.extraction)
+
+
+@agent.tool
+def expense_check(ctx: RunContext[ValidatorState]) -> AuditFinding:
+    return check_expense_totals(ctx.deps.extraction)
+
+
+@agent.tool
+def fundraising_alignment_check(ctx: RunContext[ValidatorState]) -> AuditFinding:
+    return check_fundraising_alignment(ctx.deps.extraction)
+
+
+@agent.tool
+async def verify_ein(ctx: RunContext[ValidatorState]) -> AuditFinding:
+    ein = ctx.deps.extraction.core_organization_metadata.ein
+    exists, confidence, note = await irs_ein_lookup(ein)
+    if exists:
+        return AuditFinding(
+            check_id="irs_ein_match",
+            category="Compliance",
+            severity=Severity.PASS,
+            message="EIN confirmed against IRS index.",
+            mitigation="Document verification in the filing workpapers.",
+            confidence=confidence,
+        )
+    return AuditFinding(
+        check_id="irs_ein_match",
+        category="Compliance",
+        severity=Severity.WARNING,
+        message=f"EIN {ein} could not be confirmed. {note}",
+        mitigation="Verify the EIN against the IRS EO BMF or IRS determination letter.",
+        confidence=confidence,
+    )
+
+
+@agent.output_validator
+def finalize_report(
+    ctx: RunContext[ValidatorState],
+    report: AuditReport,
+) -> AuditReport:
+    merged_findings = _merge_findings(ctx.deps.initial_findings, report.findings)
+    overall = aggregate_findings(merged_findings)
+    sections = build_section_summaries(merged_findings)
+    overall_summary = compose_overall_summary(merged_findings)
+    metadata = ctx.deps.metadata
+    notes = report.notes
+    if notes is None and isinstance(metadata, dict) and metadata.get("source"):
+        notes = f"Reviewed data source: {metadata['source']}."
+    year: int | None = None
+    if isinstance(metadata, dict):
+        metadata_year = metadata.get("return_year")
+        if metadata_year is not None:
+            try:
+                year = int(metadata_year)
+            except (TypeError, ValueError):
+                pass
+    core = ctx.deps.extraction.core_organization_metadata
+    organisation_name = core.legal_name or report.organisation_name
+    organisation_ein = core.ein or report.organisation_ein
+    return report.model_copy(
+        update={
+            "organisation_ein": organisation_ein,
+            "organisation_name": organisation_name,
+            "year": year,
+            "findings": merged_findings,
+            "overall_severity": overall,
+            "sections": sections,
+            "overall_summary": overall_summary,
+            "notes": notes,
+        }
+    )

backend/app/agents/form_auditor/checks.py

@@ -0,0 +1,282 @@
+from __future__ import annotations
+
+from collections import Counter, defaultdict
+
+from .models import (
+    AuditFinding,
+    AuditSectionSummary,
+    ExtractedIrsForm990PfDataSchema,
+    Severity,
+)
+
+
+def aggregate_findings(findings: list[AuditFinding]) -> Severity:
+    order = {Severity.ERROR: 3, Severity.WARNING: 2, Severity.PASS: 1}
+    overall = Severity.PASS
+    for finding in findings:
+        if order[finding.severity] > order[overall]:
+            overall = finding.severity
+    return overall
+
+
+def check_revenue_totals(data: ExtractedIrsForm990PfDataSchema) -> AuditFinding:
+    subtotal = sum(
+        value
+        for key, value in data.revenue_breakdown.model_dump().items()
+        if key != "total_revenue"
+    )
+    if abs(subtotal - data.revenue_breakdown.total_revenue) <= 1:
+        return AuditFinding(
+            check_id="revenue_totals",
+            category="Revenue",
+            severity=Severity.PASS,
+            message=f"Revenue categories sum (${subtotal:,.2f}) matches total revenue.",
+            mitigation="Maintain detailed support for each revenue source to preserve reconciliation trail.",
+            confidence=0.95,
+        )
+    return AuditFinding(
+        check_id="revenue_totals",
+        category="Revenue",
+        severity=Severity.ERROR,
+        message=(
+            f"Revenue categories sum (${subtotal:,.2f}) does not equal reported total "
+            f"(${data.revenue_breakdown.total_revenue:,.2f})."
+        ),
+        mitigation="Recalculate revenue totals and correct line items or Schedule A before filing.",
+        confidence=0.95,
+    )
+
+
+def check_expense_totals(data: ExtractedIrsForm990PfDataSchema) -> AuditFinding:
+    subtotal = (
+        data.expenses_breakdown.program_services_expenses
+        + data.expenses_breakdown.management_general_expenses
+        + data.expenses_breakdown.fundraising_expenses
+    )
+    if abs(subtotal - data.expenses_breakdown.total_expenses) <= 1:
+        return AuditFinding(
+            check_id="expense_totals",
+            category="Expenses",
+            severity=Severity.PASS,
+            message="Functional expenses match total expenses.",
+            mitigation="Keep functional allocation workpapers to support the reconciliation.",
+            confidence=0.95,
+        )
+    return AuditFinding(
+        check_id="expense_totals",
+        category="Expenses",
+        severity=Severity.ERROR,
+        message=(
+            f"Functional expenses (${subtotal:,.2f}) do not reconcile to total expenses "
+            f"(${data.expenses_breakdown.total_expenses:,.2f})."
+        ),
+        mitigation="Review Part I, lines 23–27 and reclassify functional expenses to tie to Part II totals.",
+        confidence=0.95,
+    )
+
+
+def check_fundraising_alignment(
+    data: ExtractedIrsForm990PfDataSchema,
+) -> AuditFinding:
+    reported_fundraising = data.expenses_breakdown.fundraising_expenses
+    event_expenses = data.fundraising_grantmaking.total_fundraising_event_expenses
+    difference = abs(reported_fundraising - event_expenses)
+    if difference <= 1:
+        return AuditFinding(
+            check_id="fundraising_alignment",
+            category="Fundraising",
+            severity=Severity.PASS,
+            message="Fundraising functional expenses align with reported event expenses.",
+            mitigation="Retain event ledgers and allocations to support matching totals.",
+            confidence=0.9,
+        )
+    severity = (
+        Severity.WARNING
+        if reported_fundraising and difference <= reported_fundraising * 0.1
+        else Severity.ERROR
+    )
+    return AuditFinding(
+        check_id="fundraising_alignment",
+        category="Fundraising",
+        severity=severity,
+        message=(
+            f"Fundraising functional expenses (${reported_fundraising:,.2f}) differ from "
+            f"reported event expenses (${event_expenses:,.2f}) by ${difference:,.2f}."
+        ),
+        mitigation="Reconcile Schedule G and Part I allocations to eliminate the variance.",
+        confidence=0.85,
+    )
+
+
+def check_balance_sheet_presence(
+    data: ExtractedIrsForm990PfDataSchema,
+) -> AuditFinding:
+    if data.balance_sheet:
+        return AuditFinding(
+            check_id="balance_sheet_present",
+            category="Balance Sheet",
+            severity=Severity.PASS,
+            message="Balance sheet data is present.",
+            mitigation="Ensure ending net assets tie to Part I, line 30.",
+            confidence=0.7,
+        )
+    return AuditFinding(
+        check_id="balance_sheet_absent",
+        category="Balance Sheet",
+        severity=Severity.WARNING,
+        message="Balance sheet section is empty; confirm Part II filing requirements.",
+        mitigation="Populate assets, liabilities, and net assets or attach supporting schedules.",
+        confidence=0.6,
+    )
+
+
+def check_governance_policies(
+    data: ExtractedIrsForm990PfDataSchema,
+) -> list[AuditFinding]:
+    gm = data.governance_management_disclosure
+    findings: list[AuditFinding] = []
+    policy_fields = {
+        "conflict_of_interest_policy": "Document the policy in Part VI or adopt one prior to filing.",
+        "whistleblower_policy": "Document whistleblower protections for staff and volunteers.",
+        "document_retention_policy": "Adopt and document a record retention policy.",
+    }
+    affirmative_fields = {
+        "financial_statements_reviewed": "Capture whether the board reviewed or audited year-end financials.",
+        "form_990_provided_to_governing_body": "Provide Form 990 to the board before submission and note the date of review.",
+    }
+
+    for field, mitigation in policy_fields.items():
+        value = (getattr(gm, field) or "").strip()
+        if not value or value.lower() in {"no", "n", "false"}:
+            findings.append(
+                AuditFinding(
+                    check_id=f"{field}_missing",
+                    category="Governance",
+                    severity=Severity.WARNING,
+                    message=f"{field.replace('_', ' ').title()} not reported or marked 'No'.",
+                    mitigation=mitigation,
+                    confidence=0.55,
+                )
+            )
+
+    for field, mitigation in affirmative_fields.items():
+        value = (getattr(gm, field) or "").strip()
+        if not value:
+            findings.append(
+                AuditFinding(
+                    check_id=f"{field}_blank",
+                    category="Governance",
+                    severity=Severity.WARNING,
+                    message=f"{field.replace('_', ' ').title()} left blank.",
+                    mitigation=mitigation,
+                    confidence=0.5,
+                )
+            )
+    return findings
+
+
+def check_board_engagement(data: ExtractedIrsForm990PfDataSchema) -> AuditFinding:
+    hours = [
+        member.average_hours_per_week
+        for member in data.officers_directors_trustees_key_employees
+        if member.average_hours_per_week is not None
+    ]
+    total_hours = sum(hours)
+    if total_hours >= 5:
+        return AuditFinding(
+            check_id="board_hours",
+            category="Governance",
+            severity=Severity.PASS,
+            message="Officer and director time commitments appear reasonable.",
+            mitigation="Continue documenting board attendance and oversight responsibilities.",
+            confidence=0.7,
+        )
+    return AuditFinding(
+        check_id="board_hours",
+        category="Governance",
+        severity=Severity.WARNING,
+        message=(
+            f"Aggregate reported board hours ({total_hours:.1f} per week) are low; "
+            "confirm entries reflect actual governance involvement."
+        ),
+        mitigation="Verify hours in Part VII; update if officers volunteer significant time.",
+        confidence=0.6,
+    )
+
+
+def check_missing_operational_details(
+    data: ExtractedIrsForm990PfDataSchema,
+) -> AuditFinding:
+    descriptors = (
+        data.functional_operational_data.fundraising_method_descriptions or ""
+    ).strip()
+    if descriptors:
+        return AuditFinding(
+            check_id="fundraising_methods_documented",
+            category="Operations",
+            severity=Severity.PASS,
+            message="Fundraising method descriptions provided.",
+            mitigation="Update narratives annually to reflect any new campaigns or joint ventures.",
+            confidence=0.65,
+        )
+    return AuditFinding(
+        check_id="fundraising_methods_missing",
+        category="Operations",
+        severity=Severity.WARNING,
+        message="Fundraising method descriptions are blank.",
+        mitigation="Add a brief Schedule G narrative describing major fundraising approaches.",
+        confidence=0.55,
+    )
+
+
+def build_section_summaries(findings: list[AuditFinding]) -> list[AuditSectionSummary]:
+    grouped: defaultdict[str, list[AuditFinding]] = defaultdict(list)
+    for finding in findings:
+        grouped[finding.category].append(finding)
+
+    summaries: list[AuditSectionSummary] = []
+    severity_order = {Severity.ERROR: 3, Severity.WARNING: 2, Severity.PASS: 1}
+    for category, category_findings in grouped.items():
+        counter = Counter(f.severity for f in category_findings)
+        severity = aggregate_findings(category_findings)
+        summary = ", ".join(
+            f"{count} {label}"
+            for label, count in (
+                ("passes", counter.get(Severity.PASS, 0)),
+                ("warnings", counter.get(Severity.WARNING, 0)),
+                ("errors", counter.get(Severity.ERROR, 0)),
+            )
+        )
+        summary_text = f"{category} review: {summary}."
+        confidence = sum(f.confidence for f in category_findings) / len(
+            category_findings
+        )
+        summaries.append(
+            AuditSectionSummary(
+                section=category,
+                severity=severity,
+                summary=summary_text,
+                confidence=confidence,
+            )
+        )
+    summaries.sort(key=lambda s: (-severity_order[s.severity], s.section.lower()))
+    return summaries
+
+
+def compose_overall_summary(findings: list[AuditFinding]) -> str:
+    if not findings:
+        return "No automated findings generated."
+    counter = Counter(f.severity for f in findings)
+    parts = []
+    if counter.get(Severity.ERROR):
+        parts.append(f"{counter[Severity.ERROR]} error(s)")
+    if counter.get(Severity.WARNING):
+        parts.append(f"{counter[Severity.WARNING]} warning(s)")
+    if counter.get(Severity.PASS):
+        parts.append(f"{counter[Severity.PASS]} check(s) passed")
+    summary = "Overall results: " + ", ".join(parts) + "."
+    return summary
+
+
+async def irs_ein_lookup(_ein: str) -> tuple[bool, float, str]:
+    return False, 0.2, "IRS verification unavailable in current environment."

backend/app/agents/form_auditor/cli.py

@@ -0,0 +1,38 @@
+from __future__ import annotations
+
+import argparse
+import asyncio
+import json
+from pathlib import Path
+
+from . import build_audit_report
+
+__all__ = ["build_audit_report", "main"]
+
+
+def _load_payload(path: Path) -> dict:
+    text = path.read_text(encoding="utf-8")
+    return json.loads(text)
+
+
+def _print_report(report: dict) -> None:
+    print(json.dumps(report, indent=2))
+
+
+def main(argv: list[str] | None = None) -> None:
+    parser = argparse.ArgumentParser(
+        description="Validate a Form 990 extraction payload using the Form Auditor agent."
+    )
+    parser.add_argument(
+        "payload",
+        nargs="?",
+        default="example_data.json",
+        help="Path to a JSON file containing the extraction payload.",
+    )
+    args = parser.parse_args(argv)
+
+    payload_path = Path(args.payload).expanduser()
+    payload = _load_payload(payload_path)
+
+    report = asyncio.run(build_audit_report(payload))
+    _print_report(report.model_dump())

backend/app/agents/form_auditor/models.py

@@ -0,0 +1,797 @@
+from __future__ import annotations
+
+import re
+from enum import Enum
+from typing import Any
+
+from pydantic import BaseModel, Field, model_validator
+
+
+class Severity(str, Enum):
+    PASS = "Pass"
+    WARNING = "Warning"
+    ERROR = "Error"
+
+
+class AuditFinding(BaseModel):
+    check_id: str
+    category: str
+    severity: Severity
+    message: str
+    mitigation: str | None = None
+    confidence: float = Field(ge=0.0, le=1.0)
+
+
+class AuditSectionSummary(BaseModel):
+    section: str
+    severity: Severity
+    summary: str
+    confidence: float = Field(ge=0.0, le=1.0)
+
+
+class AuditReport(BaseModel):
+    organisation_ein: str
+    organisation_name: str
+    year: int | None
+    overall_severity: Severity
+    findings: list[AuditFinding]
+    sections: list[AuditSectionSummary] = Field(default_factory=list)
+    overall_summary: str | None = None
+    notes: str | None = None
+
+
+class CoreOrgMetadata(BaseModel):
+    ein: str
+    legal_name: str
+    return_type: str
+    accounting_method: str
+    incorporation_state: str | None = None
+
+
+class CoreOrganizationMetadata(BaseModel):
+    ein: str = Field(
+        ...,
+        description="Unique IRS identifier for the organization.",
+        title="Employer Identification Number (EIN)",
+    )
+    legal_name: str = Field(
+        ...,
+        description="Official registered name of the organization.",
+        title="Legal Name of Organization",
+    )
+    phone_number: str = Field(
+        ..., description="Primary contact phone number.", title="Phone Number"
+    )
+    website_url: str = Field(
+        ..., description="Organization's website address.", title="Website URL"
+    )
+    return_type: str = Field(
+        ...,
+        description="Type of IRS return filed (e.g., 990, 990-EZ, 990-PF).",
+        title="Return Type",
+    )
+    amended_return: str = Field(
+        ...,
+        description="Indicates if the return is amended.",
+        title="Amended Return Flag",
+    )
+    group_exemption_number: str = Field(
+        ...,
+        description="IRS group exemption number, if applicable.",
+        title="Group Exemption Number",
+    )
+    subsection_code: str = Field(
+        ...,
+        description="IRS subsection code (e.g., 501(c)(3)).",
+        title="Subsection Code",
+    )
+    ruling_date: str = Field(
+        ...,
+        description="Date of IRS ruling or determination letter.",
+        title="Ruling/Determination Letter Date",
+    )
+    accounting_method: str = Field(
+        ...,
+        description="Accounting method used (cash, accrual, other).",
+        title="Accounting Method",
+    )
+    organization_type: str = Field(
+        ...,
+        description="Legal structure (corporation, trust, association, etc.).",
+        title="Organization Type",
+    )
+    year_of_formation: str = Field(
+        ..., description="Year the organization was formed.", title="Year of Formation"
+    )
+    incorporation_state: str = Field(
+        ..., description="State of incorporation.", title="Incorporation State"
+    )
+    calendar_year: str | None = Field(
+        default=None,
+        description="Calendar year covered by the return (if different from fiscal year).",
+        title="Calendar Year",
+    )
+
+
+class RevenueBreakdown(BaseModel):
+    total_revenue: float = Field(
+        ..., description="Sum of all revenue sources.", title="Total Revenue"
+    )
+    contributions_gifts_grants: float = Field(
+        ...,
+        description="Revenue from donations and grants.",
+        title="Contributions, Gifts, and Grants",
+    )
+    program_service_revenue: float = Field(
+        ...,
+        description="Revenue from program services.",
+        title="Program Service Revenue",
+    )
+    membership_dues: float = Field(
+        ..., description="Revenue from membership dues.", title="Membership Dues"
+    )
+    investment_income: float = Field(
+        ...,
+        description="Revenue from interest and dividends.",
+        title="Investment Income",
+    )
+    gains_losses_sales_assets: float = Field(
+        ...,
+        description="Net gains or losses from asset sales.",
+        title="Gains/Losses from Sales of Assets",
+    )
+    rental_income: float = Field(
+        ...,
+        description="Income from rental of real estate or equipment.",
+        title="Rental Income",
+    )
+    related_organizations_revenue: float = Field(
+        ...,
+        description="Revenue from related organizations.",
+        title="Related Organizations Revenue",
+    )
+    gaming_revenue: float = Field(
+        ..., description="Revenue from gaming activities.", title="Gaming Revenue"
+    )
+    other_revenue: float = Field(
+        ..., description="Miscellaneous revenue sources.", title="Other Revenue"
+    )
+    government_grants: float = Field(
+        ...,
+        description="Revenue from government grants.",
+        title="Revenue from Government Grants",
+    )
+    foreign_contributions: float = Field(
+        ..., description="Revenue from foreign sources.", title="Foreign Contributions"
+    )
+
+
+class ExpensesBreakdown(BaseModel):
+    total_expenses: float = Field(
+        ..., description="Sum of all expenses.", title="Total Functional Expenses"
+    )
+    program_services_expenses: float = Field(
+        ...,
+        description="Expenses for program services.",
+        title="Program Services Expenses",
+    )
+    management_general_expenses: float = Field(
+        ...,
+        description="Administrative and management expenses.",
+        title="Management & General Expenses",
+    )
+    fundraising_expenses: float = Field(
+        ...,
+        description="Expenses for fundraising activities.",
+        title="Fundraising Expenses",
+    )
+    grants_us_organizations: float = Field(
+        ...,
+        description="Grants and assistance to U.S. organizations.",
+        title="Grants to U.S. Organizations",
+    )
+    grants_us_individuals: float = Field(
+        ...,
+        description="Grants and assistance to U.S. individuals.",
+        title="Grants to U.S. Individuals",
+    )
+    grants_foreign_organizations: float = Field(
+        ...,
+        description="Grants and assistance to foreign organizations.",
+        title="Grants to Foreign Organizations",
+    )
+    grants_foreign_individuals: float = Field(
+        ...,
+        description="Grants and assistance to foreign individuals.",
+        title="Grants to Foreign Individuals",
+    )
+    compensation_officers: float = Field(
+        ...,
+        description="Compensation paid to officers and key employees.",
+        title="Compensation of Officers/Key Employees",
+    )
+    compensation_other_staff: float = Field(
+        ...,
+        description="Compensation paid to other staff.",
+        title="Compensation of Other Staff",
+    )
+    payroll_taxes_benefits: float = Field(
+        ...,
+        description="Payroll taxes and employee benefits.",
+        title="Payroll Taxes and Benefits",
+    )
+    professional_fees: float = Field(
+        ...,
+        description="Legal, accounting, and lobbying fees.",
+        title="Professional Fees",
+    )
+    office_occupancy_costs: float = Field(
+        ...,
+        description="Office and occupancy expenses.",
+        title="Office and Occupancy Costs",
+    )
+    information_technology_costs: float = Field(
+        ..., description="IT-related expenses.", title="Information Technology Costs"
+    )
+    travel_conference_expenses: float = Field(
+        ...,
+        description="Travel and conference costs.",
+        title="Travel and Conference Expenses",
+    )
+    depreciation_amortization: float = Field(
+        ...,
+        description="Depreciation and amortization expenses.",
+        title="Depreciation and Amortization",
+    )
+    insurance: float = Field(..., description="Insurance expenses.", title="Insurance")
+
+
+class OfficersDirectorsTrusteesKeyEmployee(BaseModel):
+    name: str = Field(..., description="Full name of the individual.", title="Name")
+    title_position: str = Field(
+        ..., description="Role or position held.", title="Title/Position"
+    )
+    average_hours_per_week: float = Field(
+        ...,
+        description="Average weekly hours devoted to position.",
+        title="Average Hours Per Week",
+    )
+    related_party_transactions: str = Field(
+        ...,
+        description="Indicates if related-party transactions occurred.",
+        title="Related-Party Transactions",
+    )
+    former_officer: str = Field(
+        ...,
+        description="Indicates if the individual is a former officer.",
+        title="Former Officer Indicator",
+    )
+    governance_role: str = Field(
+        ...,
+        description="Role in governance (voting, independent, etc.).",
+        title="Governance Role",
+    )
+
+
+class GovernanceManagementDisclosure(BaseModel):
+    governing_body_size: float = Field(
+        ...,
+        description="Number of voting members on the governing body.",
+        title="Governing Body Size",
+    )
+    independent_members: float = Field(
+        ...,
+        description="Number of independent voting members.",
+        title="Number of Independent Members",
+    )
+    financial_statements_reviewed: str = Field(
+        ...,
+        description="Indicates if financial statements were reviewed or audited.",
+        title="Financial Statements Reviewed/Audited",
+    )
+    form_990_provided_to_governing_body: str = Field(
+        ...,
+        description="Indicates if Form 990 was provided to governing body before filing.",
+        title="Form 990 Provided to Governing Body",
+    )
+    conflict_of_interest_policy: str = Field(
+        ...,
+        description="Indicates if a conflict-of-interest policy is in place.",
+        title="Conflict-of-Interest Policy",
+    )
+    whistleblower_policy: str = Field(
+        ...,
+        description="Indicates if a whistleblower policy is in place.",
+        title="Whistleblower Policy",
+    )
+    document_retention_policy: str = Field(
+        ...,
+        description="Indicates if a document retention/destruction policy is in place.",
+        title="Document Retention/Destruction Policy",
+    )
+    ceo_compensation_review_process: str = Field(
+        ...,
+        description="Description of CEO compensation review process.",
+        title="CEO Compensation Review Process",
+    )
+    public_disclosure_practices: str = Field(
+        ...,
+        description="Description of public disclosure practices.",
+        title="Public Disclosure Practices",
+    )
+
+
+class ProgramServiceAccomplishment(BaseModel):
+    program_name: str = Field(
+        ..., description="Name of the program.", title="Program Name"
+    )
+    program_description: str = Field(
+        ..., description="Description of the program.", title="Program Description"
+    )
+    expenses: float = Field(
+        ..., description="Expenses for the program.", title="Program Expenses"
+    )
+    grants: float = Field(
+        ..., description="Grants made under the program.", title="Program Grants"
+    )
+    revenue_generated: float = Field(
+        ..., description="Revenue generated by the program.", title="Revenue Generated"
+    )
+    quantitative_outputs: str = Field(
+        ...,
+        description="Quantitative outputs (e.g., number served, events held).",
+        title="Quantitative Outputs",
+    )
+
+
+class FundraisingGrantmaking(BaseModel):
+    total_fundraising_event_revenue: float = Field(
+        ...,
+        description="Total revenue from fundraising events.",
+        title="Total Fundraising Event Revenue",
+    )
+    total_fundraising_event_expenses: float = Field(
+        ...,
+        description="Total direct expenses for fundraising events.",
+        title="Total Fundraising Event Expenses",
+    )
+    professional_fundraiser_fees: float = Field(
+        ...,
+        description="Fees paid to professional fundraisers.",
+        title="Professional Fundraiser Fees",
+    )
+
+
+class FunctionalOperationalData(BaseModel):
+    number_of_employees: float = Field(
+        ..., description="Total number of employees.", title="Number of Employees"
+    )
+    number_of_volunteers: float = Field(
+        ..., description="Total number of volunteers.", title="Number of Volunteers"
+    )
+    occupancy_costs: float = Field(
+        ..., description="Total occupancy costs.", title="Occupancy Costs"
+    )
+    fundraising_method_descriptions: str = Field(
+        ...,
+        description="Descriptions of fundraising methods used.",
+        title="Fundraising Method Descriptions",
+    )
+    joint_ventures_disregarded_entities: str = Field(
+        ...,
+        description="Details of joint ventures and disregarded entities.",
+        title="Joint Ventures and Disregarded Entities",
+    )
+
+
+class CompensationDetails(BaseModel):
+    base_compensation: float = Field(
+        ..., description="Base salary or wages.", title="Base Compensation"
+    )
+    bonus: float = Field(
+        ..., description="Bonus or incentive compensation.", title="Bonus Compensation"
+    )
+    incentive: float = Field(
+        ..., description="Incentive compensation.", title="Incentive Compensation"
+    )
+    other: float = Field(
+        ..., description="Other forms of compensation.", title="Other Compensation"
+    )
+    non_fixed_compensation: str = Field(
+        ...,
+        description="Indicates if compensation is non-fixed.",
+        title="Non-Fixed Compensation Flag",
+    )
+    first_class_travel: str = Field(
+        ...,
+        description="Indicates if first-class travel was provided.",
+        title="First-Class Travel",
+    )
+    housing_allowance: str = Field(
+        ...,
+        description="Indicates if housing allowance was provided.",
+        title="Housing Allowance",
+    )
+    expense_account_usage: str = Field(
+        ...,
+        description="Indicates if expense account was used.",
+        title="Expense Account Usage",
+    )
+    supplemental_retirement: str = Field(
+        ...,
+        description="Indicates if supplemental retirement or deferred comp was provided.",
+        title="Supplemental Retirement/Deferred Comp",
+    )
+
+
+class PoliticalLobbyingActivities(BaseModel):
+    lobbying_expenditures_direct: float = Field(
+        ...,
+        description="Direct lobbying expenditures.",
+        title="Direct Lobbying Expenditures",
+    )
+    lobbying_expenditures_grassroots: float = Field(
+        ...,
+        description="Grassroots lobbying expenditures.",
+        title="Grassroots Lobbying Expenditures",
+    )
+    election_501h_status: str = Field(
+        ...,
+        description="Indicates if 501(h) election was made.",
+        title="501(h) Election Status",
+    )
+    political_campaign_expenditures: float = Field(
+        ...,
+        description="Expenditures for political campaigns.",
+        title="Political Campaign Expenditures",
+    )
+    related_organizations_affiliates: str = Field(
+        ...,
+        description="Details of related organizations or affiliates involved.",
+        title="Related Organizations/Affiliates Involved",
+    )
+
+
+class InvestmentsEndowment(BaseModel):
+    investment_types: str = Field(
+        ...,
+        description="Types of investments held (securities, partnerships, real estate).",
+        title="Investment Types",
+    )
+    donor_restricted_endowment_values: float = Field(
+        ...,
+        description="Value of donor-restricted endowments.",
+        title="Donor-Restricted Endowment Values",
+    )
+    net_appreciation_depreciation: float = Field(
+        ...,
+        description="Net appreciation or depreciation of investments.",
+        title="Net Appreciation/Depreciation",
+    )
+    related_organization_transactions: str = Field(
+        ...,
+        description="Details of transactions with related organizations.",
+        title="Related Organization Transactions",
+    )
+    loans_to_from_related_parties: str = Field(
+        ...,
+        description="Details of loans to or from related parties.",
+        title="Loans to/from Related Parties",
+    )
+
+
+class TaxCompliancePenalties(BaseModel):
+    penalties_excise_taxes_reported: str = Field(
+        ...,
+        description="Reported penalties or excise taxes.",
+        title="Penalties or Excise Taxes Reported",
+    )
+    unrelated_business_income_disclosure: str = Field(
+        ...,
+        description="Disclosure of unrelated business income (UBI).",
+        title="Unrelated Business Income Disclosure",
+    )
+    foreign_bank_account_reporting: str = Field(
+        ...,
+        description="Disclosure of foreign bank accounts (FBAR equivalent).",
+        title="Foreign Bank Account Reporting",
+    )
+    schedule_o_narrative_explanations: str = Field(
+        ...,
+        description="Narrative explanations from Schedule O.",
+        title="Schedule O Narrative Explanations",
+    )
+
+
+_OFFICER_HOURS_PATTERN = re.compile(r"([\d.]+)\s*hrs?/wk", re.IGNORECASE)
+
+
+def _parse_officer_list(entries: list[str] | None) -> list[dict[str, Any]]:
+    if not entries:
+        return []
+    parsed: list[dict[str, Any]] = []
+    for raw in entries:
+        if not isinstance(raw, str):
+            continue
+        parts = [part.strip() for part in raw.split(",")]
+        name = parts[0] if parts else ""
+        title = parts[1] if len(parts) > 1 else ""
+        role = parts[3] if len(parts) > 3 else ""
+        hours = 0.0
+        match = _OFFICER_HOURS_PATTERN.search(raw)
+        if match:
+            try:
+                hours = float(match.group(1))
+            except ValueError:
+                hours = 0.0
+        parsed.append(
+            {
+                "name": name,
+                "title_position": title,
+                "average_hours_per_week": hours,
+                "related_party_transactions": "",
+                "former_officer": "",
+                "governance_role": role,
+            }
+        )
+    return parsed
+
+
+def _build_program_accomplishments(
+    descriptions: list[str] | None,
+) -> list[dict[str, Any]]:
+    if not descriptions:
+        return []
+    programs: list[dict[str, Any]] = []
+    for idx, description in enumerate(descriptions, start=1):
+        if not isinstance(description, str):
+            continue
+        programs.append(
+            {
+                "program_name": f"Program {idx}",
+                "program_description": description.strip(),
+                "expenses": 0.0,
+                "grants": 0.0,
+                "revenue_generated": 0.0,
+                "quantitative_outputs": "",
+            }
+        )
+    return programs
+
+
+def _transform_flat_payload(data: dict[str, Any]) -> dict[str, Any]:
+    def get_str(key: str) -> str:
+        value = data.get(key)
+        if value is None:
+            return ""
+        return str(value)
+
+    def get_value(key: str, default: Any = 0) -> Any:
+        return data.get(key, default)
+
+    transformed: dict[str, Any] = {
+        "core_organization_metadata": {
+            "ein": get_str("ein"),
+            "legal_name": get_str("legal_name"),
+            "phone_number": get_str("phone_number"),
+            "website_url": get_str("website_url"),
+            "return_type": get_str("return_type"),
+            "amended_return": get_str("amended_return"),
+            "group_exemption_number": get_str("group_exemption_number"),
+            "subsection_code": get_str("subsection_code"),
+            "ruling_date": get_str("ruling_date"),
+            "accounting_method": get_str("accounting_method"),
+            "organization_type": get_str("organization_type"),
+            "year_of_formation": get_str("year_of_formation"),
+            "incorporation_state": get_str("incorporation_state"),
+            "calendar_year": get_str("calendar_year"),
+        },
+        "revenue_breakdown": {
+            "total_revenue": get_value("total_revenue"),
+            "contributions_gifts_grants": get_value("contributions_gifts_grants"),
+            "program_service_revenue": get_value("program_service_revenue"),
+            "membership_dues": get_value("membership_dues"),
+            "investment_income": get_value("investment_income"),
+            "gains_losses_sales_assets": get_value("gains_losses_sales_assets"),
+            "rental_income": get_value("rental_income"),
+            "related_organizations_revenue": get_value("related_organizations_revenue"),
+            "gaming_revenue": get_value("gaming_revenue"),
+            "other_revenue": get_value("other_revenue"),
+            "government_grants": get_value("government_grants"),
+            "foreign_contributions": get_value("foreign_contributions"),
+        },
+        "expenses_breakdown": {
+            "total_expenses": get_value("total_expenses"),
+            "program_services_expenses": get_value("program_services_expenses"),
+            "management_general_expenses": get_value("management_general_expenses"),
+            "fundraising_expenses": get_value("fundraising_expenses"),
+            "grants_us_organizations": get_value("grants_us_organizations"),
+            "grants_us_individuals": get_value("grants_us_individuals"),
+            "grants_foreign_organizations": get_value("grants_foreign_organizations"),
+            "grants_foreign_individuals": get_value("grants_foreign_individuals"),
+            "compensation_officers": get_value("compensation_officers"),
+            "compensation_other_staff": get_value("compensation_other_staff"),
+            "payroll_taxes_benefits": get_value("payroll_taxes_benefits"),
+            "professional_fees": get_value("professional_fees"),
+            "office_occupancy_costs": get_value("office_occupancy_costs"),
+            "information_technology_costs": get_value("information_technology_costs"),
+            "travel_conference_expenses": get_value("travel_conference_expenses"),
+            "depreciation_amortization": get_value("depreciation_amortization"),
+            "insurance": get_value("insurance"),
+        },
+        "balance_sheet": data.get("balance_sheet") or {},
+        "officers_directors_trustees_key_employees": _parse_officer_list(
+            data.get("officers_list")
+        ),
+        "governance_management_disclosure": {
+            "governing_body_size": get_value("governing_body_size"),
+            "independent_members": get_value("independent_members"),
+            "financial_statements_reviewed": get_str("financial_statements_reviewed"),
+            "form_990_provided_to_governing_body": get_str(
+                "form_990_provided_to_governing_body"
+            ),
+            "conflict_of_interest_policy": get_str("conflict_of_interest_policy"),
+            "whistleblower_policy": get_str("whistleblower_policy"),
+            "document_retention_policy": get_str("document_retention_policy"),
+            "ceo_compensation_review_process": get_str(
+                "ceo_compensation_review_process"
+            ),
+            "public_disclosure_practices": get_str("public_disclosure_practices"),
+        },
+        "program_service_accomplishments": _build_program_accomplishments(
+            data.get("program_accomplishments_list")
+        ),
+        "fundraising_grantmaking": {
+            "total_fundraising_event_revenue": get_value(
+                "total_fundraising_event_revenue"
+            ),
+            "total_fundraising_event_expenses": get_value(
+                "total_fundraising_event_expenses"
+            ),
+            "professional_fundraiser_fees": get_value("professional_fundraiser_fees"),
+        },
+        "functional_operational_data": {
+            "number_of_employees": get_value("number_of_employees"),
+            "number_of_volunteers": get_value("number_of_volunteers"),
+            "occupancy_costs": get_value("occupancy_costs"),
+            "fundraising_method_descriptions": get_str(
+                "fundraising_method_descriptions"
+            ),
+            "joint_ventures_disregarded_entities": get_str(
+                "joint_ventures_disregarded_entities"
+            ),
+        },
+        "compensation_details": {
+            "base_compensation": get_value("base_compensation"),
+            "bonus": get_value("bonus"),
+            "incentive": get_value("incentive"),
+            "other": get_value("other_compensation", get_value("other", 0)),
+            "non_fixed_compensation": get_str("non_fixed_compensation"),
+            "first_class_travel": get_str("first_class_travel"),
+            "housing_allowance": get_str("housing_allowance"),
+            "expense_account_usage": get_str("expense_account_usage"),
+            "supplemental_retirement": get_str("supplemental_retirement"),
+        },
+        "political_lobbying_activities": {
+            "lobbying_expenditures_direct": get_value("lobbying_expenditures_direct"),
+            "lobbying_expenditures_grassroots": get_value(
+                "lobbying_expenditures_grassroots"
+            ),
+            "election_501h_status": get_str("election_501h_status"),
+            "political_campaign_expenditures": get_value(
+                "political_campaign_expenditures"
+            ),
+            "related_organizations_affiliates": get_str(
+                "related_organizations_affiliates"
+            ),
+        },
+        "investments_endowment": {
+            "investment_types": get_str("investment_types"),
+            "donor_restricted_endowment_values": get_value(
+                "donor_restricted_endowment_values"
+            ),
+            "net_appreciation_depreciation": get_value("net_appreciation_depreciation"),
+            "related_organization_transactions": get_str(
+                "related_organization_transactions"
+            ),
+            "loans_to_from_related_parties": get_str("loans_to_from_related_parties"),
+        },
+        "tax_compliance_penalties": {
+            "penalties_excise_taxes_reported": get_str(
+                "penalties_excise_taxes_reported"
+            ),
+            "unrelated_business_income_disclosure": get_str(
+                "unrelated_business_income_disclosure"
+            ),
+            "foreign_bank_account_reporting": get_str("foreign_bank_account_reporting"),
+            "schedule_o_narrative_explanations": get_str(
+                "schedule_o_narrative_explanations"
+            ),
+        },
+    }
+    return transformed
+
+
+class ExtractedIrsForm990PfDataSchema(BaseModel):
+    core_organization_metadata: CoreOrganizationMetadata = Field(
+        ...,
+        description="Essential identifiers and attributes for normalizing entities across filings and years.",
+        title="Core Organization Metadata",
+    )
+    revenue_breakdown: RevenueBreakdown = Field(
+        ...,
+        description="Detailed breakdown of revenue streams for the fiscal year.",
+        title="Revenue Breakdown",
+    )
+    expenses_breakdown: ExpensesBreakdown = Field(
+        ...,
+        description="Detailed breakdown of expenses for the fiscal year.",
+        title="Expenses Breakdown",
+    )
+    balance_sheet: dict[str, Any] = Field(
+        default_factory=dict,
+        description="Assets, liabilities, and net assets at year end.",
+        title="Balance Sheet Data",
+    )
+    officers_directors_trustees_key_employees: list[
+        OfficersDirectorsTrusteesKeyEmployee
+    ] = Field(
+        ...,
+        description="List of key personnel and their compensation.",
+        title="Officers, Directors, Trustees, Key Employees",
+    )
+    governance_management_disclosure: GovernanceManagementDisclosure = Field(
+        ...,
+        description="Governance and management practices, policies, and disclosures.",
+        title="Governance, Management, and Disclosure",
+    )
+    program_service_accomplishments: list[ProgramServiceAccomplishment] = Field(
+        ...,
+        description="Major programs and their outputs for the fiscal year.",
+        title="Program Service Accomplishments",
+    )
+    fundraising_grantmaking: FundraisingGrantmaking = Field(
+        ...,
+        description="Fundraising event details and grantmaking activities.",
+        title="Fundraising & Grantmaking",
+    )
+    functional_operational_data: FunctionalOperationalData = Field(
+        ...,
+        description="Operational metrics and related-organization relationships.",
+        title="Functional & Operational Data",
+    )
+    compensation_details: CompensationDetails = Field(
+        ...,
+        description="Detailed breakdown of officer compensation and benefits.",
+        title="Compensation Details",
+    )
+    political_lobbying_activities: PoliticalLobbyingActivities = Field(
+        ...,
+        description="Details of political and lobbying expenditures and affiliations.",
+        title="Political & Lobbying Activities",
+    )
+    investments_endowment: InvestmentsEndowment = Field(
+        ...,
+        description="Investment holdings, endowment values, and related transactions.",
+        title="Investments & Endowment",
+    )
+    tax_compliance_penalties: TaxCompliancePenalties = Field(
+        ...,
+        description="Tax compliance indicators, penalties, and narrative explanations.",
+        title="Tax Compliance / Penalties",
+    )
+
+    @model_validator(mode="before")
+    @classmethod
+    def _ensure_structure(cls, value: Any) -> Any:
+        if not isinstance(value, dict):
+            return value
+        if "core_organization_metadata" in value:
+            return value
+        return _transform_flat_payload(value)
+
+
+class ValidatorState(BaseModel):
+    extraction: ExtractedIrsForm990PfDataSchema
+    initial_findings: list[AuditFinding] = Field(default_factory=list)
+    metadata: dict[str, Any] = Field(default_factory=dict)

backend/app/agents/web_search/__init__.py

@@ -0,0 +1,37 @@
+from __future__ import annotations
+
+from .agent import agent
+from .models import WebSearchResponse, WebSearchState
+
+
+async def search_web(
+    query: str,
+    max_results: int = 5,
+    include_raw_content: bool = False,
+) -> WebSearchResponse:
+    """
+    Execute web search using Tavily MCP server.
+
+    Args:
+        query: Search query string
+        max_results: Maximum number of results to return (1-10)
+        include_raw_content: Whether to include full content in results
+
+    Returns:
+        WebSearchResponse with results and summary
+    """
+    state = WebSearchState(
+        user_query=query,
+        max_results=max_results,
+        include_raw_content=include_raw_content,
+    )
+
+    prompt = (
+        f"Search the web for: {query}\n\n"
+        f"Return the top {max_results} most relevant results. "
+        "Provide a concise summary of the key findings."
+    )
+
+    # Ejecutar agente con Tavily API directa
+    result = await agent.run(prompt, deps=state)
+    return result.output

backend/app/agents/web_search/agent.py

@@ -0,0 +1,72 @@
+from __future__ import annotations
+
+from pydantic_ai import Agent, RunContext
+from pydantic_ai.models.openai import OpenAIChatModel
+from pydantic_ai.providers.azure import AzureProvider
+from tavily import TavilyClient
+
+from app.core.config import settings
+
+from .models import WebSearchResponse, WebSearchState, SearchResult
+
+
+provider = AzureProvider(
+    azure_endpoint=settings.AZURE_OPENAI_ENDPOINT,
+    api_version=settings.AZURE_OPENAI_API_VERSION,
+    api_key=settings.AZURE_OPENAI_API_KEY,
+)
+model = OpenAIChatModel(model_name="gpt-4o", provider=provider)
+
+
+tavily_client = TavilyClient(api_key=settings.TAVILY_API_KEY)
+
+agent = Agent(
+    model=model,
+    name="WebSearchAgent",
+    deps_type=WebSearchState,
+    output_type=WebSearchResponse,
+    system_prompt=(
+        "You are a web search assistant powered by Tavily. "
+        "Use the tavily_search tool to find relevant, up-to-date information. "
+        "Return a structured WebSearchResponse with results and a concise summary. "
+        "Always cite your sources with URLs."
+    ),
+)
+
+
+@agent.tool
+def tavily_search(ctx: RunContext[WebSearchState], query: str) -> list[SearchResult]:
+    """Search the web using Tavily API for up-to-date information."""
+    response = tavily_client.search(
+        query=query,
+        max_results=ctx.deps.max_results,
+        search_depth="basic",
+        include_raw_content=ctx.deps.include_raw_content,
+    )
+
+    results = []
+    for item in response.get("results", []):
+        results.append(
+            SearchResult(
+                title=item.get("title", ""),
+                url=item.get("url", ""),
+                content=item.get("content", ""),
+                score=item.get("score"),
+            )
+        )
+
+    return results
+
+
+@agent.output_validator
+def finalize_response(
+    ctx: RunContext[WebSearchState],
+    response: WebSearchResponse,
+) -> WebSearchResponse:
+    """Post-process and validate the search response"""
+    return response.model_copy(
+        update={
+            "query": ctx.deps.user_query,
+            "total_results": len(response.results),
+        }
+    )

backend/app/agents/web_search/models.py

@@ -0,0 +1,29 @@
+from __future__ import annotations
+
+from pydantic import BaseModel, Field
+
+
+class SearchResult(BaseModel):
+    """Individual search result from web search"""
+
+    title: str
+    url: str
+    content: str
+    score: float | None = None
+
+
+class WebSearchState(BaseModel):
+    """State passed to agent tools via deps"""
+
+    user_query: str
+    max_results: int = Field(default=5, ge=1, le=10)
+    include_raw_content: bool = False
+
+
+class WebSearchResponse(BaseModel):
+    """Structured response from the web search agent"""
+
+    query: str
+    results: list[SearchResult]
+    summary: str
+    total_results: int

backend/app/core/config.py

@@ -37,10 +37,17 @@ class Settings(BaseSettings):
     # Azure OpenAI configuración
     AZURE_OPENAI_ENDPOINT: str
     AZURE_OPENAI_API_KEY: str
-    AZURE_OPENAI_API_VERSION: str = "2024-02-01"
+    AZURE_OPENAI_API_VERSION: str = "2024-08-01-preview"
     AZURE_OPENAI_EMBEDDING_MODEL: str = "text-embedding-3-large"
     AZURE_OPENAI_EMBEDDING_DEPLOYMENT: str = "text-embedding-3-large"
 
+    # Rate limiting para embeddings (ajustar según tier de Azure OpenAI)
+    # S0 tier: batch_size=16, delay=1.0 es seguro
+    # Tier superior: batch_size=100, delay=0.1
+    EMBEDDING_BATCH_SIZE: int = 16
+    EMBEDDING_DELAY_BETWEEN_BATCHES: float = 1.0
+    EMBEDDING_MAX_RETRIES: int = 5
+
     # Google Cloud / Vertex AI configuración
     GOOGLE_APPLICATION_CREDENTIALS: str
     GOOGLE_CLOUD_PROJECT: str
@@ -51,6 +58,9 @@ class Settings(BaseSettings):
     LANDINGAI_API_KEY: str
     LANDINGAI_ENVIRONMENT: str = "production"  # "production" o "eu"
     
+    TAVILY_API_KEY: str
+    
+
     # Schemas storage
     SCHEMAS_DIR: str = "./data/schemas"
 

backend/app/example_data.json

@@ -0,0 +1,608 @@
+{
+  "extraction": {
+    "core_organization_metadata": {
+      "ein": "84-2674654",
+      "legal_name": "07 IN HEAVEN MEMORIAL SCHOLARSHIP",
+      "phone_number": "(262) 215-0300",
+      "website_url": "",
+      "return_type": "990-PF",
+      "amended_return": "No",
+      "group_exemption_number": "",
+      "subsection_code": "501(c)(3)",
+      "ruling_date": "",
+      "accounting_method": "Cash",
+      "organization_type": "corporation",
+      "year_of_formation": "",
+      "incorporation_state": "WI"
+    },
+    "revenue_breakdown": {
+      "total_revenue": 5227,
+      "contributions_gifts_grants": 5227,
+      "program_service_revenue": 0,
+      "membership_dues": 0,
+      "investment_income": 0,
+      "gains_losses_sales_assets": 0,
+      "rental_income": 0,
+      "related_organizations_revenue": 0,
+      "gaming_revenue": 0,
+      "other_revenue": 0,
+      "government_grants": 0,
+      "foreign_contributions": 0
+    },
+    "expenses_breakdown": {
+      "total_expenses": 2104,
+      "program_services_expenses": 0,
+      "management_general_expenses": 0,
+      "fundraising_expenses": 2104,
+      "grants_us_organizations": 0,
+      "grants_us_individuals": 0,
+      "grants_foreign_organizations": 0,
+      "grants_foreign_individuals": 0,
+      "compensation_officers": 0,
+      "compensation_other_staff": 0,
+      "payroll_taxes_benefits": 0,
+      "professional_fees": 0,
+      "office_occupancy_costs": 0,
+      "information_technology_costs": 0,
+      "travel_conference_expenses": 0,
+      "depreciation_amortization": 0,
+      "insurance": 0
+    },
+    "balance_sheet": {},
+    "officers_directors_trustees_key_employees": [
+      {
+        "name": "REBECCA TERPSTRA",
+        "title_position": "PRESIDENT",
+        "average_hours_per_week": 0.1,
+        "related_party_transactions": "",
+        "former_officer": "",
+        "governance_role": ""
+      },
+      {
+        "name": "ROBERT GUZMAN",
+        "title_position": "VICE PRESDEINT",
+        "average_hours_per_week": 0.1,
+        "related_party_transactions": "",
+        "former_officer": "",
+        "governance_role": ""
+      },
+      {
+        "name": "ANDREA VALENTI",
+        "title_position": "TREASURER",
+        "average_hours_per_week": 0.1,
+        "related_party_transactions": "",
+        "former_officer": "",
+        "governance_role": ""
+      },
+      {
+        "name": "BETHANY WALSH",
+        "title_position": "SECRETARY",
+        "average_hours_per_week": 0.1,
+        "related_party_transactions": "",
+        "former_officer": "",
+        "governance_role": ""
+      }
+    ],
+    "governance_management_disclosure": {
+      "governing_body_size": 4,
+      "independent_members": 4,
+      "financial_statements_reviewed": "",
+      "form_990_provided_to_governing_body": "",
+      "conflict_of_interest_policy": "",
+      "whistleblower_policy": "",
+      "document_retention_policy": "",
+      "ceo_compensation_review_process": "",
+      "public_disclosure_practices": "Yes"
+    },
+    "program_service_accomplishments": [],
+    "fundraising_grantmaking": {
+      "total_fundraising_event_revenue": 0,
+      "total_fundraising_event_expenses": 2104,
+      "professional_fundraiser_fees": 0
+    },
+    "functional_operational_data": {
+      "number_of_employees": 0,
+      "number_of_volunteers": 0,
+      "occupancy_costs": 0,
+      "fundraising_method_descriptions": "",
+      "joint_ventures_disregarded_entities": ""
+    },
+    "compensation_details": {
+      "base_compensation": 0,
+      "bonus": 0,
+      "incentive": 0,
+      "other": 0,
+      "non_fixed_compensation": "",
+      "first_class_travel": "",
+      "housing_allowance": "",
+      "expense_account_usage": "",
+      "supplemental_retirement": ""
+    },
+    "political_lobbying_activities": {
+      "lobbying_expenditures_direct": 0,
+      "lobbying_expenditures_grassroots": 0,
+      "election_501h_status": "",
+      "political_campaign_expenditures": 0,
+      "related_organizations_affiliates": ""
+    },
+    "investments_endowment": {
+      "investment_types": "",
+      "donor_restricted_endowment_values": 0,
+      "net_appreciation_depreciation": 0,
+      "related_organization_transactions": "",
+      "loans_to_from_related_parties": ""
+    },
+    "tax_compliance_penalties": {
+      "penalties_excise_taxes_reported": "No",
+      "unrelated_business_income_disclosure": "No",
+      "foreign_bank_account_reporting": "No",
+      "schedule_o_narrative_explanations": ""
+    }
+  },
+  "extraction_metadata": {
+    "core_organization_metadata": {
+      "ein": {
+        "value": "84-2674654",
+        "references": ["0-7"]
+      },
+      "legal_name": {
+        "value": "07 IN HEAVEN MEMORIAL SCHOLARSHIP",
+        "references": ["0-6"]
+      },
+      "phone_number": {
+        "value": "(262) 215-0300",
+        "references": ["0-a"]
+      },
+      "website_url": {
+        "value": "",
+        "references": []
+      },
+      "return_type": {
+        "value": "990-PF",
+        "references": ["4ade8ed0-bce7-4bd5-bd8d-190e3e4be95b"]
+      },
+      "amended_return": {
+        "value": "No",
+        "references": ["4ac9edc4-e9bb-430f-b4c4-a42bf4c04b28"]
+      },
+      "group_exemption_number": {
+        "value": "",
+        "references": []
+      },
+      "subsection_code": {
+        "value": "501(c)(3)",
+        "references": ["4ac9edc4-e9bb-430f-b4c4-a42bf4c04b28"]
+      },
+      "ruling_date": {
+        "value": "",
+        "references": []
+      },
+      "accounting_method": {
+        "value": "Cash",
+        "references": ["0-d"]
+      },
+      "organization_type": {
+        "value": "corporation",
+        "references": ["4ac9edc4-e9bb-430f-b4c4-a42bf4c04b28"]
+      },
+      "year_of_formation": {
+        "value": "",
+        "references": []
+      },
+      "incorporation_state": {
+        "value": "WI",
+        "references": ["4ac9edc4-e9bb-430f-b4c4-a42bf4c04b28"]
+      }
+    },
+    "revenue_breakdown": {
+      "total_revenue": {
+        "value": 5227,
+        "references": ["0-1z"]
+      },
+      "contributions_gifts_grants": {
+        "value": 5227,
+        "references": ["0-m"]
+      },
+      "program_service_revenue": {
+        "value": 0,
+        "references": []
+      },
+      "membership_dues": {
+        "value": 0,
+        "references": []
+      },
+      "investment_income": {
+        "value": 0,
+        "references": []
+      },
+      "gains_losses_sales_assets": {
+        "value": 0,
+        "references": []
+      },
+      "rental_income": {
+        "value": 0,
+        "references": []
+      },
+      "related_organizations_revenue": {
+        "value": 0,
+        "references": []
+      },
+      "gaming_revenue": {
+        "value": 0,
+        "references": []
+      },
+      "other_revenue": {
+        "value": 0,
+        "references": []
+      },
+      "government_grants": {
+        "value": 0,
+        "references": []
+      },
+      "foreign_contributions": {
+        "value": 0,
+        "references": []
+      }
+    },
+    "expenses_breakdown": {
+      "total_expenses": {
+        "value": 2104,
+        "references": ["0-2S"]
+      },
+      "program_services_expenses": {
+        "value": 0,
+        "references": []
+      },
+      "management_general_expenses": {
+        "value": 0,
+        "references": []
+      },
+      "fundraising_expenses": {
+        "value": 2104,
+        "references": ["13-d"]
+      },
+      "grants_us_organizations": {
+        "value": 0,
+        "references": []
+      },
+      "grants_us_individuals": {
+        "value": 0,
+        "references": []
+      },
+      "grants_foreign_organizations": {
+        "value": 0,
+        "references": []
+      },
+      "grants_foreign_individuals": {
+        "value": 0,
+        "references": []
+      },
+      "compensation_officers": {
+        "value": 0,
+        "references": ["5-1q", "5-1w", "5-1C", "5-1I"]
+      },
+      "compensation_other_staff": {
+        "value": 0,
+        "references": []
+      },
+      "payroll_taxes_benefits": {
+        "value": 0,
+        "references": []
+      },
+      "professional_fees": {
+        "value": 0,
+        "references": []
+      },
+      "office_occupancy_costs": {
+        "value": 0,
+        "references": []
+      },
+      "information_technology_costs": {
+        "value": 0,
+        "references": []
+      },
+      "travel_conference_expenses": {
+        "value": 0,
+        "references": []
+      },
+      "depreciation_amortization": {
+        "value": 0,
+        "references": []
+      },
+      "insurance": {
+        "value": 0,
+        "references": []
+      }
+    },
+    "balance_sheet": {},
+    "officers_directors_trustees_key_employees": [
+      {
+        "name": {
+          "value": "REBECCA TERPSTRA",
+          "references": ["5-1o"]
+        },
+        "title_position": {
+          "value": "PRESIDENT",
+          "references": ["5-1p"]
+        },
+        "average_hours_per_week": {
+          "value": 0.1,
+          "references": ["5-1p"]
+        },
+        "related_party_transactions": {
+          "value": "",
+          "references": []
+        },
+        "former_officer": {
+          "value": "",
+          "references": []
+        },
+        "governance_role": {
+          "value": "",
+          "references": []
+        }
+      },
+      {
+        "name": {
+          "value": "ROBERT GUZMAN",
+          "references": ["5-1u"]
+        },
+        "title_position": {
+          "value": "VICE PRESDEINT",
+          "references": ["5-1v"]
+        },
+        "average_hours_per_week": {
+          "value": 0.1,
+          "references": ["5-1v"]
+        },
+        "related_party_transactions": {
+          "value": "",
+          "references": []
+        },
+        "former_officer": {
+          "value": "",
+          "references": []
+        },
+        "governance_role": {
+          "value": "",
+          "references": []
+        }
+      },
+      {
+        "name": {
+          "value": "ANDREA VALENTI",
+          "references": ["5-1A"]
+        },
+        "title_position": {
+          "value": "TREASURER",
+          "references": ["5-1B"]
+        },
+        "average_hours_per_week": {
+          "value": 0.1,
+          "references": ["5-1B"]
+        },
+        "related_party_transactions": {
+          "value": "",
+          "references": []
+        },
+        "former_officer": {
+          "value": "",
+          "references": []
+        },
+        "governance_role": {
+          "value": "",
+          "references": []
+        }
+      },
+      {
+        "name": {
+          "value": "BETHANY WALSH",
+          "references": ["5-1G"]
+        },
+        "title_position": {
+          "value": "SECRETARY",
+          "references": ["5-1H"]
+        },
+        "average_hours_per_week": {
+          "value": 0.1,
+          "references": ["5-1H"]
+        },
+        "related_party_transactions": {
+          "value": "",
+          "references": []
+        },
+        "former_officer": {
+          "value": "",
+          "references": []
+        },
+        "governance_role": {
+          "value": "",
+          "references": []
+        }
+      }
+    ],
+    "governance_management_disclosure": {
+      "governing_body_size": {
+        "value": 4,
+        "references": ["5-1o", "5-1u", "5-1A", "5-1G"]
+      },
+      "independent_members": {
+        "value": 4,
+        "references": ["5-1o", "5-1u", "5-1A", "5-1G"]
+      },
+      "financial_statements_reviewed": {
+        "value": "",
+        "references": []
+      },
+      "form_990_provided_to_governing_body": {
+        "value": "",
+        "references": []
+      },
+      "conflict_of_interest_policy": {
+        "value": "",
+        "references": []
+      },
+      "whistleblower_policy": {
+        "value": "",
+        "references": []
+      },
+      "document_retention_policy": {
+        "value": "",
+        "references": []
+      },
+      "ceo_compensation_review_process": {
+        "value": "",
+        "references": []
+      },
+      "public_disclosure_practices": {
+        "value": "Yes",
+        "references": ["4-g"]
+      }
+    },
+    "program_service_accomplishments": [],
+    "fundraising_grantmaking": {
+      "total_fundraising_event_revenue": {
+        "value": 0,
+        "references": []
+      },
+      "total_fundraising_event_expenses": {
+        "value": 2104,
+        "references": ["13-d"]
+      },
+      "professional_fundraiser_fees": {
+        "value": 0,
+        "references": []
+      }
+    },
+    "functional_operational_data": {
+      "number_of_employees": {
+        "value": 0,
+        "references": []
+      },
+      "number_of_volunteers": {
+        "value": 0,
+        "references": []
+      },
+      "occupancy_costs": {
+        "value": 0,
+        "references": []
+      },
+      "fundraising_method_descriptions": {
+        "value": "",
+        "references": []
+      },
+      "joint_ventures_disregarded_entities": {
+        "value": "",
+        "references": []
+      }
+    },
+    "compensation_details": {
+      "base_compensation": {
+        "value": 0,
+        "references": ["5-1q", "5-1w"]
+      },
+      "bonus": {
+        "value": 0,
+        "references": []
+      },
+      "incentive": {
+        "value": 0,
+        "references": []
+      },
+      "other": {
+        "value": 0,
+        "references": []
+      },
+      "non_fixed_compensation": {
+        "value": "",
+        "references": []
+      },
+      "first_class_travel": {
+        "value": "",
+        "references": []
+      },
+      "housing_allowance": {
+        "value": "",
+        "references": []
+      },
+      "expense_account_usage": {
+        "value": "",
+        "references": []
+      },
+      "supplemental_retirement": {
+        "value": "",
+        "references": []
+      }
+    },
+    "political_lobbying_activities": {
+      "lobbying_expenditures_direct": {
+        "value": 0,
+        "references": []
+      },
+      "lobbying_expenditures_grassroots": {
+        "value": 0,
+        "references": []
+      },
+      "election_501h_status": {
+        "value": "",
+        "references": []
+      },
+      "political_campaign_expenditures": {
+        "value": 0,
+        "references": []
+      },
+      "related_organizations_affiliates": {
+        "value": "",
+        "references": []
+      }
+    },
+    "investments_endowment": {
+      "investment_types": {
+        "value": "",
+        "references": []
+      },
+      "donor_restricted_endowment_values": {
+        "value": 0,
+        "references": []
+      },
+      "net_appreciation_depreciation": {
+        "value": 0,
+        "references": []
+      },
+      "related_organization_transactions": {
+        "value": "",
+        "references": []
+      },
+      "loans_to_from_related_parties": {
+        "value": "",
+        "references": []
+      }
+    },
+    "tax_compliance_penalties": {
+      "penalties_excise_taxes_reported": {
+        "value": "No",
+        "references": ["3-I"]
+      },
+      "unrelated_business_income_disclosure": {
+        "value": "No",
+        "references": ["3-Y"]
+      },
+      "foreign_bank_account_reporting": {
+        "value": "No",
+        "references": ["4-H"]
+      },
+      "schedule_o_narrative_explanations": {
+        "value": "",
+        "references": []
+      }
+    }
+  },
+  "metadata": {
+    "filename": "markdown.md",
+    "org_id": null,
+    "duration_ms": 16656,
+    "credit_usage": 27.2,
+    "job_id": "nnmr8lcxtykk5ll5wodjtrnn6",
+    "version": "extract-20250930"
+  }
+}

backend/app/main.py

@@ -11,6 +11,7 @@ from .routers.agent import router as agent_router
 from .routers.chunking import router as chunking_router
 from .routers.chunking_landingai import router as chunking_landingai_router
 from .routers.dataroom import router as dataroom_router
+from .routers.extracted_data import router as extracted_data_router
 from .routers.files import router as files_router
 from .routers.schemas import router as schemas_router
 from .routers.vectors import router as vectors_router
@@ -123,6 +124,9 @@ app.include_router(schemas_router)
 # Chunking LandingAI router (nuevo)
 app.include_router(chunking_landingai_router)
 
+# Extracted data router (nuevo)
+app.include_router(extracted_data_router)
+
 app.include_router(dataroom_router, prefix="/api/v1")
 
 app.include_router(agent_router)

backend/app/models/extracted_data.py

@@ -0,0 +1,68 @@
+"""
+Modelo Redis-OM para almacenar datos extraídos de documentos.
+Permite búsqueda rápida de datos estructurados sin necesidad de búsqueda vectorial.
+"""
+from datetime import datetime
+from typing import Optional, Dict, Any
+from redis_om import HashModel, Field, Migrator
+import json
+
+
+class ExtractedDocument(HashModel):
+    """
+    Modelo para guardar datos extraídos de documentos en Redis.
+
+    Uso:
+    1. Cuando se procesa un PDF con schema y se extraen datos
+    2. Los chunks van a Qdrant (para RAG)
+    3. Los datos extraídos van a Redis (para búsqueda estructurada)
+
+    Ventajas:
+    - Búsqueda rápida por file_name, tema, collection_name
+    - Acceso directo a datos extraídos sin búsqueda vectorial
+    - Permite filtros y agregaciones
+    """
+
+    # Identificadores
+    file_name: str = Field(index=True)
+    tema: str = Field(index=True)
+    collection_name: str = Field(index=True)
+
+    # Datos extraídos (JSON serializado)
+    # Redis-OM HashModel no soporta Dict directamente, usamos str y serializamos
+    extracted_data_json: str
+
+    # Metadata
+    extraction_timestamp: str  # ISO format
+
+    class Meta:
+        database = None  # Se configura en runtime
+        global_key_prefix = "extracted_doc"
+        model_key_prefix = "doc"
+
+    def set_extracted_data(self, data: Dict[str, Any]) -> None:
+        """Helper para serializar datos extraídos a JSON"""
+        self.extracted_data_json = json.dumps(data, ensure_ascii=False, indent=2)
+
+    def get_extracted_data(self) -> Dict[str, Any]:
+        """Helper para deserializar datos extraídos desde JSON"""
+        return json.loads(self.extracted_data_json)
+
+    @classmethod
+    def find_by_file(cls, file_name: str):
+        """Busca todos los documentos extraídos de un archivo"""
+        return cls.find(cls.file_name == file_name).all()
+
+    @classmethod
+    def find_by_tema(cls, tema: str):
+        """Busca todos los documentos extraídos de un tema"""
+        return cls.find(cls.tema == tema).all()
+
+    @classmethod
+    def find_by_collection(cls, collection_name: str):
+        """Busca todos los documentos en una colección"""
+        return cls.find(cls.collection_name == collection_name).all()
+
+
+# Ejecutar migración para crear índices en Redis
+Migrator().run()

backend/app/models/schema_models.py

@@ -58,7 +58,7 @@ class CustomSchema(BaseModel):
     schema_id: Optional[str] = Field(None, description="ID único del schema (generado automáticamente si no se provee)")
     schema_name: str = Field(..., description="Nombre descriptivo del schema", min_length=1, max_length=100)
     description: str = Field(..., description="Descripción de qué extrae este schema", min_length=1, max_length=500)
-    fields: List[SchemaField] = Field(..., description="Lista de campos a extraer", min_items=1, max_items=50)
+    fields: List[SchemaField] = Field(..., description="Lista de campos a extraer", min_items=1, max_items=200)
 
     # Metadata
     created_at: Optional[str] = Field(None, description="Timestamp de creación ISO")

backend/app/routers/agent.py

@@ -1,12 +1,19 @@
-from fastapi import APIRouter
+import json
-from pydantic_ai import Agent
+import logging
+from dataclasses import dataclass
+from typing import Annotated, Any
+
+from fastapi import APIRouter, Header
+from pydantic_ai import Agent, RunContext
 from pydantic_ai.models.openai import OpenAIChatModel
 from pydantic_ai.providers.azure import AzureProvider
 from pydantic_ai.ui.vercel_ai import VercelAIAdapter
 from starlette.requests import Request
 from starlette.responses import Response
 
+from app.agents import analyst, form_auditor, web_search
 from app.core.config import settings
+from app.services.extracted_data_service import get_extracted_data_service
 
 provider = AzureProvider(
     azure_endpoint=settings.AZURE_OPENAI_ENDPOINT,
@@ -14,11 +21,65 @@ provider = AzureProvider(
     api_key=settings.AZURE_OPENAI_API_KEY,
 )
 model = OpenAIChatModel(model_name="gpt-4o", provider=provider)
-agent = Agent(model=model)
+
+
+@dataclass
+class Deps:
+    extracted_data: list[dict[str, Any]]
+
+
+agent = Agent(model=model, deps_type=Deps)
 
 router = APIRouter(prefix="/api/v1/agent", tags=["Agent"])
 
+logger = logging.getLogger(__name__)
+
+
+@agent.tool
+async def build_audit_report(ctx: RunContext[Deps]):
+    """Calls the audit subagent to get a full audit report of the organization"""
+    data = ctx.deps.extracted_data[0]
+
+    result = await form_auditor.build_audit_report(data)
+
+    return result.model_dump()
+
+
+@agent.tool
+async def build_analysis_report(ctx: RunContext[Deps]):
+    """Calls the analyst subagent to get a full report of the organization's performance across years"""
+    data = ctx.deps.extracted_data
+    if not data:
+        raise ValueError("No extracted data available for analysis.")
+
+    if len(data) == 1:
+        logger.info(
+            "build_analysis_report called with single-year data; report will still be generated but trends may be limited."
+        )
+
+    result = await analyst.build_performance_report(data)
+
+    return result.model_dump()
+
+
+@agent.tool_plain
+async def search_web_information(query: str, max_results: int = 5):
+    """Search the web for up-to-date information using Tavily. Use this when you need current information, news, research, or facts not in your knowledge base."""
+    result = await web_search.search_web(query=query, max_results=max_results)
+
+    return result.model_dump()
+
 
 @router.post("/chat")
-async def chat(request: Request) -> Response:
+async def chat(request: Request, tema: Annotated[str, Header()]) -> Response:
-    return await VercelAIAdapter.dispatch_request(request, agent=agent)
+    extracted_data_service = get_extracted_data_service()
+
+    data = await extracted_data_service.get_by_tema(tema)
+
+    extracted_data = [doc.get_extracted_data() for doc in data]
+
+    logger.info(f"Extracted data amount: {len(extracted_data)}")
+
+    deps = Deps(extracted_data=extracted_data)
+
+    return await VercelAIAdapter.dispatch_request(request, agent=agent, deps=deps)

backend/app/routers/chunking_landingai.py

@@ -2,17 +2,19 @@
 Router para procesamiento de PDFs con LandingAI.
 Soporta dos modos: rápido (solo parse) y extracción (parse + extract con schema).
 """
+
 import logging
 import time
+from typing import List, Literal, Optional
+
 from fastapi import APIRouter, HTTPException
-from pydantic import BaseModel, Field
-from typing import Optional, List, Literal
-
 from langchain_core.documents import Document
+from pydantic import BaseModel, Field
 
-from ..services.landingai_service import get_landingai_service
-from ..services.chunking_service import get_chunking_service
 from ..repositories.schema_repository import get_schema_repository
+from ..services.chunking_service import get_chunking_service
+from ..services.landingai_service import get_landingai_service
+from ..services.extracted_data_service import get_extracted_data_service
 from ..utils.chunking.token_manager import TokenManager
 
 logger = logging.getLogger(__name__)
@@ -22,6 +24,7 @@ router = APIRouter(prefix="/api/v1/chunking-landingai", tags=["chunking-landinga
 
 class ProcessLandingAIRequest(BaseModel):
     """Request para procesar PDF con LandingAI"""
+
     file_name: str = Field(..., description="Nombre del archivo PDF")
     tema: str = Field(..., description="Tema/carpeta del archivo")
     collection_name: str = Field(..., description="Colección de Qdrant")
@@ -29,34 +32,33 @@ class ProcessLandingAIRequest(BaseModel):
     # Modo de procesamiento
     mode: Literal["quick", "extract"] = Field(
         default="quick",
-        description="Modo: 'quick' (solo parse) o 'extract' (parse + datos estructurados)"
+        description="Modo: 'quick' (solo parse) o 'extract' (parse + datos estructurados)",
     )
 
     # Schema (obligatorio si mode='extract')
     schema_id: Optional[str] = Field(
-        None,
+        None, description="ID del schema a usar (requerido si mode='extract')"
-        description="ID del schema a usar (requerido si mode='extract')"
     )
 
     # Configuración de chunks
     include_chunk_types: List[str] = Field(
         default=["text", "table"],
-        description="Tipos de chunks a incluir: text, table, figure, etc."
+        description="Tipos de chunks a incluir: text, table, figure, etc.",
     )
     max_tokens_per_chunk: int = Field(
         default=1500,
         ge=500,
         le=3000,
-        description="Tokens máximos por chunk (flexible para tablas/figuras)"
+        description="Tokens máximos por chunk (flexible para tablas/figuras)",
     )
     merge_small_chunks: bool = Field(
-        default=True,
+        default=True, description="Unir chunks pequeños de la misma página y tipo"
-        description="Unir chunks pequeños de la misma página y tipo"
     )
 
 
 class ProcessLandingAIResponse(BaseModel):
     """Response del procesamiento con LandingAI"""
+
     success: bool
     mode: str
     processing_time_seconds: float
@@ -98,20 +100,21 @@ async def process_with_landingai(request: ProcessLandingAIRequest):
 
     try:
         logger.info(f"\n{'=' * 60}")
-        logger.info(f"INICIANDO PROCESAMIENTO CON LANDINGAI")
+        logger.info("INICIANDO PROCESAMIENTO CON LANDINGAI")
         logger.info(f"{'=' * 60}")
         logger.info(f"Archivo: {request.file_name}")
         logger.info(f"Tema: {request.tema}")
         logger.info(f"Modo: {request.mode}")
         logger.info(f"Colección: {request.collection_name}")
+        logger.info(f"Schema ID recibido: '{request.schema_id}' (tipo: {type(request.schema_id).__name__})")
 
         # 1. Validar schema si es modo extract
         custom_schema = None
         if request.mode == "extract":
-            if not request.schema_id:
+            if not request.schema_id or request.schema_id.strip() == "":
                 raise HTTPException(
                     status_code=400,
-                    detail="schema_id es requerido cuando mode='extract'"
+                    detail="schema_id es requerido cuando mode='extract'",
                 )
 
             schema_repo = get_schema_repository()
@@ -119,8 +122,7 @@ async def process_with_landingai(request: ProcessLandingAIRequest):
 
             if not custom_schema:
                 raise HTTPException(
-                    status_code=404,
+                    status_code=404, detail=f"Schema no encontrado: {request.schema_id}"
-                    detail=f"Schema no encontrado: {request.schema_id}"
                 )
 
             logger.info(f"Schema seleccionado: {custom_schema.schema_name}")
@@ -131,14 +133,12 @@ async def process_with_landingai(request: ProcessLandingAIRequest):
 
         try:
             pdf_bytes = await chunking_service.download_pdf_from_blob(
-                request.file_name,
+                request.file_name, request.tema
-                request.tema
             )
         except Exception as e:
             logger.error(f"Error descargando PDF: {e}")
             raise HTTPException(
-                status_code=404,
+                status_code=404, detail=f"No se pudo descargar el PDF: {str(e)}"
-                detail=f"No se pudo descargar el PDF: {str(e)}"
             )
 
         # 3. Procesar con LandingAI
@@ -150,13 +150,12 @@ async def process_with_landingai(request: ProcessLandingAIRequest):
                 pdf_bytes=pdf_bytes,
                 file_name=request.file_name,
                 custom_schema=custom_schema,
-                include_chunk_types=request.include_chunk_types
+                include_chunk_types=request.include_chunk_types,
             )
         except Exception as e:
             logger.error(f"Error en LandingAI: {e}")
             raise HTTPException(
-                status_code=500,
+                status_code=500, detail=f"Error procesando con LandingAI: {str(e)}"
-                detail=f"Error procesando con LandingAI: {str(e)}"
             )
 
         documents = result["chunks"]
@@ -164,7 +163,7 @@ async def process_with_landingai(request: ProcessLandingAIRequest):
         if not documents:
             raise HTTPException(
                 status_code=400,
-                detail="No se generaron chunks después del procesamiento"
+                detail="No se generaron chunks después del procesamiento",
             )
 
         # 4. Aplicar control flexible de tokens
@@ -172,7 +171,7 @@ async def process_with_landingai(request: ProcessLandingAIRequest):
         documents = _apply_flexible_token_control(
             documents,
             max_tokens=request.max_tokens_per_chunk,
-            merge_small=request.merge_small_chunks
+            merge_small=request.merge_small_chunks,
         )
 
         # 5. Generar embeddings
@@ -180,13 +179,16 @@ async def process_with_landingai(request: ProcessLandingAIRequest):
         texts = [doc.page_content for doc in documents]
 
         try:
-            embeddings = await chunking_service.embedding_service.generate_embeddings_batch(texts)
+            embeddings = (
+                await chunking_service.embedding_service.generate_embeddings_batch(
+                    texts
+                )
+            )
             logger.info(f"Embeddings generados: {len(embeddings)} vectores")
         except Exception as e:
             logger.error(f"Error generando embeddings: {e}")
             raise HTTPException(
-                status_code=500,
+                status_code=500, detail=f"Error generando embeddings: {str(e)}"
-                detail=f"Error generando embeddings: {str(e)}"
             )
 
         # 6. Preparar chunks para Qdrant con IDs determinísticos
@@ -198,32 +200,48 @@ async def process_with_landingai(request: ProcessLandingAIRequest):
             chunk_id = chunking_service._generate_deterministic_id(
                 file_name=request.file_name,
                 page=doc.metadata.get("page", 1),
-                chunk_index=doc.metadata.get("chunk_id", str(idx))
+                chunk_index=doc.metadata.get("chunk_id", str(idx)),
             )
 
-            qdrant_chunks.append({
+            qdrant_chunks.append(
+                {
                     "id": chunk_id,
                     "vector": embedding,
                     "payload": {
                         "page_content": doc.page_content,
-                    "metadata": doc.metadata  # Metadata rica de LandingAI
+                        "metadata": doc.metadata,  # Metadata rica de LandingAI
+                    },
                 }
-            })
+            )
 
         # 7. Subir a Qdrant
         try:
             upload_result = await chunking_service.vector_db.add_chunks(
-                request.collection_name,
+                request.collection_name, qdrant_chunks
-                qdrant_chunks
             )
             logger.info(f"Subida completada: {upload_result['chunks_added']} chunks")
         except Exception as e:
             logger.error(f"Error subiendo a Qdrant: {e}")
             raise HTTPException(
-                status_code=500,
+                status_code=500, detail=f"Error subiendo a Qdrant: {str(e)}"
-                detail=f"Error subiendo a Qdrant: {str(e)}"
             )
 
+        # 8. Guardar datos extraídos en Redis (si existe extracted_data)
+        if result.get("extracted_data") and result["extracted_data"].get("extraction"):
+            try:
+                logger.info("\n[6/6] Guardando datos extraídos en Redis...")
+                extracted_data_service = get_extracted_data_service()
+
+                await extracted_data_service.save_extracted_data(
+                    file_name=request.file_name,
+                    tema=request.tema,
+                    collection_name=request.collection_name,
+                    extracted_data=result["extracted_data"]["extraction"]
+                )
+            except Exception as e:
+                # No fallar si Redis falla, solo logear
+                logger.warning(f"⚠️  No se pudieron guardar datos en Redis (no crítico): {e}")
+
         # Tiempo total
         processing_time = time.time() - start_time
 
@@ -245,23 +263,18 @@ async def process_with_landingai(request: ProcessLandingAIRequest):
             schema_used=custom_schema.schema_id if custom_schema else None,
             extracted_data=result.get("extracted_data"),
             parse_metadata=result["parse_metadata"],
-            message=f"PDF procesado exitosamente en modo {request.mode}"
+            message=f"PDF procesado exitosamente en modo {request.mode}",
         )
 
     except HTTPException:
         raise
     except Exception as e:
         logger.error(f"Error inesperado en procesamiento: {e}")
-        raise HTTPException(
+        raise HTTPException(status_code=500, detail=f"Error inesperado: {str(e)}")
-            status_code=500,
-            detail=f"Error inesperado: {str(e)}"
-        )
 
 
 def _apply_flexible_token_control(
-    documents: List[Document],
+    documents: List[Document], max_tokens: int, merge_small: bool
-    max_tokens: int,
-    merge_small: bool
 ) -> List[Document]:
     """
     Aplica control flexible de tokens (Opción C del diseño).
@@ -306,11 +319,7 @@ def _apply_flexible_token_control(
 
         else:
             # Intentar merge si es pequeño
-            if (
+            if merge_small and tokens < max_tokens * 0.5 and i < len(documents) - 1:
-                merge_small and
-                tokens < max_tokens * 0.5 and
-                i < len(documents) - 1
-            ):
                 next_doc = documents[i + 1]
                 if _can_merge(doc, next_doc, max_tokens, token_manager):
                     logger.debug(f"Merging chunks {i} y {i + 1}")
@@ -326,9 +335,7 @@ def _apply_flexible_token_control(
 
 
 def _split_large_chunk(
-    doc: Document,
+    doc: Document, max_tokens: int, token_manager: TokenManager
-    max_tokens: int,
-    token_manager: TokenManager
 ) -> List[Document]:
     """Divide un chunk grande en sub-chunks"""
     content = doc.page_content
@@ -343,8 +350,7 @@ def _split_large_chunk(
             # Guardar chunk actual
             sub_content = " ".join(current_chunk)
             sub_doc = Document(
-                page_content=sub_content,
+                page_content=sub_content, metadata={**doc.metadata, "is_split": True}
-                metadata={**doc.metadata, "is_split": True}
             )
             sub_chunks.append(sub_doc)
             current_chunk = [word]
@@ -357,8 +363,7 @@ def _split_large_chunk(
     if current_chunk:
         sub_content = " ".join(current_chunk)
         sub_doc = Document(
-            page_content=sub_content,
+            page_content=sub_content, metadata={**doc.metadata, "is_split": True}
-            metadata={**doc.metadata, "is_split": True}
         )
         sub_chunks.append(sub_doc)
 
@@ -366,10 +371,7 @@ def _split_large_chunk(
 
 
 def _can_merge(
-    doc1: Document,
+    doc1: Document, doc2: Document, max_tokens: int, token_manager: TokenManager
-    doc2: Document,
-    max_tokens: int,
-    token_manager: TokenManager
 ) -> bool:
     """Verifica si dos docs se pueden mergear"""
     # Misma página
@@ -391,6 +393,5 @@ def _merge_documents(doc1: Document, doc2: Document) -> Document:
     """Mergea dos documentos"""
     merged_content = f"{doc1.page_content}\n\n{doc2.page_content}"
     return Document(
-        page_content=merged_content,
+        page_content=merged_content, metadata={**doc1.metadata, "is_merged": True}
-        metadata={**doc1.metadata, "is_merged": True}
     )

backend/app/routers/dataroom.py

@@ -1,10 +1,12 @@
 import logging
+from typing import Optional
 
 from fastapi import APIRouter, HTTPException
 from pydantic import BaseModel
 
 from ..models.dataroom import DataRoom
 from ..models.vector_models import CollectionCreateRequest
+from ..services.azure_service import azure_service
 from ..services.vector_service import vector_service
 
 logger = logging.getLogger(__name__)
@@ -12,13 +14,146 @@ logger = logging.getLogger(__name__)
 
 class DataroomCreate(BaseModel):
     name: str
-    collection: str = ""
+
-    storage: str = ""
+    @property
+    def collection(self) -> str:
+        return self.name.lower().replace(" ", "_")
+
+    @property
+    def storage(self) -> str:
+        return self.name.lower().replace(" ", "_")
+
+
+class DataroomInfo(BaseModel):
+    name: str
+    collection: str
+    storage: str
+    file_count: int
+    total_size_bytes: int
+    total_size_mb: float
+    collection_exists: bool
+    vector_count: Optional[int]
+    collection_info: Optional[dict]
+    file_types: dict
+    recent_files: list
 
 
 router = APIRouter(prefix="/dataroom", tags=["Dataroom"])
 
 
+@router.get("/{dataroom_name}/info")
+async def dataroom_info(dataroom_name: str) -> DataroomInfo:
+    """
+    Obtener información detallada de un dataroom específico
+    """
+    try:
+        # Find the dataroom in Redis
+        datarooms = DataRoom.find().all()
+        dataroom = None
+        for room in datarooms:
+            if room.name == dataroom_name:
+                dataroom = room
+                break
+
+        if not dataroom:
+            raise HTTPException(
+                status_code=404, detail=f"Dataroom '{dataroom_name}' not found"
+            )
+
+        # Get file information from Azure Storage
+        try:
+            files_data = await azure_service.list_files(dataroom_name)
+        except Exception as e:
+            logger.warning(f"Could not fetch files for dataroom '{dataroom_name}': {e}")
+            files_data = []
+
+        # Calculate file metrics
+        file_count = len(files_data)
+        total_size_bytes = sum(file_data.get("size", 0) for file_data in files_data)
+        total_size_mb = (
+            round(total_size_bytes / (1024 * 1024), 2) if total_size_bytes > 0 else 0.0
+        )
+
+        # Analyze file types
+        file_types = {}
+        recent_files = []
+
+        for file_data in files_data:
+            # Count file types by extension
+            filename = file_data.get("name", "")
+            if "." in filename:
+                ext = filename.split(".")[-1].lower()
+                file_types[ext] = file_types.get(ext, 0) + 1
+
+            # Collect recent files (up to 5)
+            if len(recent_files) < 5:
+                recent_files.append(
+                    {
+                        "name": filename,
+                        "size_mb": round(file_data.get("size", 0) / (1024 * 1024), 2),
+                        "last_modified": file_data.get("last_modified"),
+                    }
+                )
+
+        # Sort recent files by last modified (newest first)
+        recent_files.sort(key=lambda x: x.get("last_modified", ""), reverse=True)
+
+        # Get vector collection information
+        collection_exists = False
+        vector_count = None
+        collection_info = None
+
+        try:
+            collection_exists_response = await vector_service.check_collection_exists(
+                dataroom_name
+            )
+            collection_exists = collection_exists_response.exists
+
+            if collection_exists:
+                collection_info_response = await vector_service.get_collection_info(
+                    dataroom_name
+                )
+                if collection_info_response:
+                    collection_info = {
+                        "vectors_count": collection_info_response.vectors_count,
+                        "indexed_vectors_count": collection_info_response.vectors_count,
+                        "points_count": collection_info_response.vectors_count,
+                        "segments_count": collection_info_response.vectors_count,
+                        "status": collection_info_response.status,
+                    }
+                    vector_count = collection_info_response.vectors_count
+        except Exception as e:
+            logger.warning(
+                f"Could not fetch collection info for '{dataroom_name}': {e}"
+            )
+
+        logger.info(
+            f"Retrieved info for dataroom '{dataroom_name}': {file_count} files, {total_size_mb}MB"
+        )
+
+        return DataroomInfo(
+            name=dataroom.name,
+            collection=dataroom.collection,
+            storage=dataroom.storage,
+            file_count=file_count,
+            total_size_bytes=total_size_bytes,
+            total_size_mb=total_size_mb,
+            collection_exists=collection_exists,
+            vector_count=vector_count,
+            collection_info=collection_info,
+            file_types=file_types,
+            recent_files=recent_files,
+        )
+
+    except HTTPException:
+        raise
+    except Exception as e:
+        logger.error(f"Error getting dataroom info for '{dataroom_name}': {e}")
+        raise HTTPException(
+            status_code=500, detail=f"Error getting dataroom info: {str(e)}"
+        )
+
+
 @router.get("/")
 async def list_datarooms():
     """

backend/app/routers/extracted_data.py

@@ -0,0 +1,141 @@
+"""
+Router para consultar datos extraídos almacenados en Redis.
+"""
+import logging
+from typing import List, Optional
+
+from fastapi import APIRouter, HTTPException, Query
+from pydantic import BaseModel
+
+from ..services.extracted_data_service import get_extracted_data_service
+
+logger = logging.getLogger(__name__)
+
+router = APIRouter(prefix="/api/v1/extracted-data", tags=["extracted-data"])
+
+
+class ExtractedDataResponse(BaseModel):
+    """Response con datos extraídos de un documento"""
+    pk: str
+    file_name: str
+    tema: str
+    collection_name: str
+    extracted_data: dict
+    extraction_timestamp: str
+
+
+class ExtractedDataListResponse(BaseModel):
+    """Response con lista de datos extraídos"""
+    total: int
+    documents: List[ExtractedDataResponse]
+
+
+@router.get("/by-file/{file_name}", response_model=ExtractedDataListResponse)
+async def get_by_file(file_name: str):
+    """
+    Obtiene todos los datos extraídos de un archivo específico.
+
+    Args:
+        file_name: Nombre del archivo
+
+    Returns:
+        Lista de documentos con datos extraídos
+    """
+    try:
+        service = get_extracted_data_service()
+        docs = await service.get_by_file(file_name)
+
+        documents = [
+            ExtractedDataResponse(
+                pk=doc.pk,
+                file_name=doc.file_name,
+                tema=doc.tema,
+                collection_name=doc.collection_name,
+                extracted_data=doc.get_extracted_data(),
+                extraction_timestamp=doc.extraction_timestamp
+            )
+            for doc in docs
+        ]
+
+        return ExtractedDataListResponse(
+            total=len(documents),
+            documents=documents
+        )
+
+    except Exception as e:
+        logger.error(f"Error obteniendo datos extraídos por archivo: {e}")
+        raise HTTPException(status_code=500, detail=str(e))
+
+
+@router.get("/by-tema/{tema}", response_model=ExtractedDataListResponse)
+async def get_by_tema(tema: str):
+    """
+    Obtiene todos los datos extraídos de un tema específico.
+
+    Args:
+        tema: Nombre del tema
+
+    Returns:
+        Lista de documentos con datos extraídos
+    """
+    try:
+        service = get_extracted_data_service()
+        docs = await service.get_by_tema(tema)
+
+        documents = [
+            ExtractedDataResponse(
+                pk=doc.pk,
+                file_name=doc.file_name,
+                tema=doc.tema,
+                collection_name=doc.collection_name,
+                extracted_data=doc.get_extracted_data(),
+                extraction_timestamp=doc.extraction_timestamp
+            )
+            for doc in docs
+        ]
+
+        return ExtractedDataListResponse(
+            total=len(documents),
+            documents=documents
+        )
+
+    except Exception as e:
+        logger.error(f"Error obteniendo datos extraídos por tema: {e}")
+        raise HTTPException(status_code=500, detail=str(e))
+
+
+@router.get("/by-collection/{collection_name}", response_model=ExtractedDataListResponse)
+async def get_by_collection(collection_name: str):
+    """
+    Obtiene todos los datos extraídos de una colección específica.
+
+    Args:
+        collection_name: Nombre de la colección
+
+    Returns:
+        Lista de documentos con datos extraídos
+    """
+    try:
+        service = get_extracted_data_service()
+        docs = await service.get_by_collection(collection_name)
+
+        documents = [
+            ExtractedDataResponse(
+                pk=doc.pk,
+                file_name=doc.file_name,
+                tema=doc.tema,
+                collection_name=doc.collection_name,
+                extracted_data=doc.get_extracted_data(),
+                extraction_timestamp=doc.extraction_timestamp
+            )
+            for doc in docs
+        ]
+
+        return ExtractedDataListResponse(
+            total=len(documents),
+            documents=documents
+        )
+
+    except Exception as e:
+        logger.error(f"Error obteniendo datos extraídos por colección: {e}")
+        raise HTTPException(status_code=500, detail=str(e))

backend/app/services/chunking_service.py

@@ -66,6 +66,8 @@ class ChunkingService:
         """
         Descarga un PDF desde Azure Blob Storage.
 
+        NOTA: Todos los blobs se guardan en minúsculas en Azure.
+
         Args:
             file_name: Nombre del archivo
             tema: Tema/carpeta del archivo
@@ -77,8 +79,9 @@ class ChunkingService:
             Exception: Si hay error descargando el archivo
         """
         try:
-            blob_path = f"{tema}/{file_name}"
+            # Convertir a minúsculas ya que todos los blobs están en minúsculas
-            logger.info(f"Descargando PDF: {blob_path}")
+            blob_path = f"{tema.lower()}/{file_name.lower()}"
+            logger.info(f"Descargando PDF: {blob_path} (tema original: {tema}, file original: {file_name})")
 
             blob_client = self.blob_service.get_blob_client(
                 container=self.container_name,

backend/app/services/embedding_service.py

@@ -1,10 +1,12 @@
 """
 Servicio de embeddings usando Azure OpenAI.
 Genera embeddings para chunks de texto usando text-embedding-3-large (3072 dimensiones).
+Incluye manejo de rate limits con retry exponencial y delays entre batches.
 """
+import asyncio
 import logging
 from typing import List
-from openai import AzureOpenAI
+from openai import AzureOpenAI, RateLimitError
 from ..core.config import settings
 
 logger = logging.getLogger(__name__)
@@ -63,28 +65,47 @@ class EmbeddingService:
     async def generate_embeddings_batch(
         self,
         texts: List[str],
-        batch_size: int = 100
+        batch_size: int | None = None,
+        delay_between_batches: float | None = None,
+        max_retries: int | None = None
     ) -> List[List[float]]:
         """
-        Genera embeddings para múltiples textos en lotes.
+        Genera embeddings para múltiples textos en lotes con manejo de rate limits.
 
         Args:
             texts: Lista de textos para generar embeddings
-            batch_size: Tamaño del lote para procesamiento (default: 100)
+            batch_size: Tamaño del lote (None = usar configuración de settings)
+            delay_between_batches: Segundos de espera entre batches (None = usar configuración)
+            max_retries: Número máximo de reintentos (None = usar configuración)
 
         Returns:
             Lista de vectores de embeddings
 
         Raises:
-            Exception: Si hay error al generar los embeddings
+            Exception: Si hay error al generar los embeddings después de todos los reintentos
         """
+        # Usar configuración de settings si no se proporciona
+        batch_size = batch_size or settings.EMBEDDING_BATCH_SIZE
+        delay_between_batches = delay_between_batches or settings.EMBEDDING_DELAY_BETWEEN_BATCHES
+        max_retries = max_retries or settings.EMBEDDING_MAX_RETRIES
+
         try:
             embeddings = []
+            total_batches = (len(texts) - 1) // batch_size + 1
+
+            logger.info(f"Iniciando generación de embeddings: {len(texts)} textos en {total_batches} batches")
+            logger.info(f"Configuración: batch_size={batch_size}, delay={delay_between_batches}s, max_retries={max_retries}")
 
             for i in range(0, len(texts), batch_size):
                 batch = texts[i:i + batch_size]
-                logger.info(f"Procesando lote {i//batch_size + 1}/{(len(texts)-1)//batch_size + 1}")
+                batch_num = i // batch_size + 1
 
+                logger.info(f"📊 Procesando batch {batch_num}/{total_batches} ({len(batch)} textos)...")
+
+                # Retry con exponential backoff
+                retry_count = 0
+                while retry_count <= max_retries:
+                    try:
                         response = self.client.embeddings.create(
                             input=batch,
                             model=self.model
@@ -101,8 +122,32 @@ class EmbeddingService:
                                 )
 
                         embeddings.extend(batch_embeddings)
+                        logger.info(f"✓ Batch {batch_num}/{total_batches} completado exitosamente")
+                        break  # Éxito, salir del retry loop
 
-            logger.info(f"Generados {len(embeddings)} embeddings exitosamente")
+                    except RateLimitError as e:
+                        retry_count += 1
+                        if retry_count > max_retries:
+                            logger.error(f"❌ Rate limit excedido después de {max_retries} reintentos")
+                            raise
+
+                        # Exponential backoff: 2^retry_count segundos
+                        wait_time = 2 ** retry_count
+                        logger.warning(
+                            f"⚠️  Rate limit alcanzado en batch {batch_num}/{total_batches}. "
+                            f"Reintento {retry_count}/{max_retries} en {wait_time}s..."
+                        )
+                        await asyncio.sleep(wait_time)
+
+                    except Exception as e:
+                        logger.error(f"❌ Error en batch {batch_num}/{total_batches}: {e}")
+                        raise
+
+                # Delay entre batches para respetar rate limit (excepto en el último)
+                if i + batch_size < len(texts):
+                    await asyncio.sleep(delay_between_batches)
+
+            logger.info(f"✅ Embeddings generados exitosamente: {len(embeddings)} vectores de {self.embedding_dimension}D")
             return embeddings
 
         except Exception as e:

backend/app/services/extracted_data_service.py

@@ -0,0 +1,131 @@
+"""
+Servicio para manejar el almacenamiento de datos extraídos en Redis.
+"""
+import logging
+from datetime import datetime
+from typing import Dict, Any, List, Optional
+
+from ..models.extracted_data import ExtractedDocument
+
+logger = logging.getLogger(__name__)
+
+
+class ExtractedDataService:
+    """Servicio para guardar y recuperar datos extraídos de documentos"""
+
+    async def save_extracted_data(
+        self,
+        file_name: str,
+        tema: str,
+        collection_name: str,
+        extracted_data: Dict[str, Any]
+    ) -> ExtractedDocument:
+        """
+        Guarda datos extraídos de un documento en Redis.
+
+        Args:
+            file_name: Nombre del archivo
+            tema: Tema del documento
+            collection_name: Colección de Qdrant
+            extracted_data: Datos extraídos (dict)
+
+        Returns:
+            ExtractedDocument guardado
+        """
+        try:
+            # Crear instancia del modelo
+            doc = ExtractedDocument(
+                file_name=file_name,
+                tema=tema,
+                collection_name=collection_name,
+                extracted_data_json="",  # Se setea después
+                extraction_timestamp=datetime.utcnow().isoformat()
+            )
+
+            # Serializar datos extraídos
+            doc.set_extracted_data(extracted_data)
+
+            # Guardar en Redis
+            doc.save()
+
+            logger.info(
+                f"💾 Datos extraídos guardados en Redis: {file_name} "
+                f"({len(extracted_data)} campos)"
+            )
+
+            return doc
+
+        except Exception as e:
+            logger.error(f"Error guardando datos extraídos en Redis: {e}")
+            raise
+
+    async def get_by_file(self, file_name: str) -> List[ExtractedDocument]:
+        """
+        Obtiene todos los documentos extraídos de un archivo.
+
+        Args:
+            file_name: Nombre del archivo
+
+        Returns:
+            Lista de ExtractedDocument
+        """
+        try:
+            docs = ExtractedDocument.find_by_file(file_name)
+            logger.info(f"Encontrados {len(docs)} documentos extraídos para {file_name}")
+            return docs
+        except Exception as e:
+            logger.error(f"Error buscando documentos por archivo: {e}")
+            return []
+
+    async def get_by_tema(self, tema: str) -> List[ExtractedDocument]:
+        """
+        Obtiene todos los documentos extraídos de un tema.
+
+        Args:
+            tema: Tema a buscar
+
+        Returns:
+            Lista de ExtractedDocument
+        """
+        try:
+            docs = ExtractedDocument.find_by_tema(tema)
+            logger.info(f"Encontrados {len(docs)} documentos extraídos para tema {tema}")
+            return docs
+        except Exception as e:
+            logger.error(f"Error buscando documentos por tema: {e}")
+            return []
+
+    async def get_by_collection(self, collection_name: str) -> List[ExtractedDocument]:
+        """
+        Obtiene todos los documentos de una colección.
+
+        Args:
+            collection_name: Nombre de la colección
+
+        Returns:
+            Lista de ExtractedDocument
+        """
+        try:
+            docs = ExtractedDocument.find_by_collection(collection_name)
+            logger.info(f"Encontrados {len(docs)} documentos en colección {collection_name}")
+            return docs
+        except Exception as e:
+            logger.error(f"Error buscando documentos por colección: {e}")
+            return []
+
+
+# Instancia global singleton
+_extracted_data_service: Optional[ExtractedDataService] = None
+
+
+def get_extracted_data_service() -> ExtractedDataService:
+    """
+    Obtiene la instancia singleton del servicio.
+
+    Returns:
+        Instancia de ExtractedDataService
+    """
+    global _extracted_data_service
+    if _extracted_data_service is None:
+        _extracted_data_service = ExtractedDataService()
+    return _extracted_data_service

backend/data/schemas/schema_103b7090a542.json

@@ -0,0 +1,776 @@
+{
+  "schema_id": "schema_103b7090a542",
+  "schema_name": "Form 990-PF Data Extraction",
+  "description": "Comprehensive data extraction schema for IRS Form 990-PF (Private Foundation) including financial, governance, and operational information",
+  "fields": [
+    {
+      "name": "ein",
+      "type": "string",
+      "description": "Federal Employer Identification Number of the organization",
+      "required": true,
+      "min_value": null,
+      "max_value": null,
+      "pattern": "^\\d{2}-\\d{7}$"
+    },
+    {
+      "name": "calendar_year",
+      "type": "integer",
+      "description": "Calendar year for which the data is reported",
+      "required": true,
+      "min_value": null,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "legal_name",
+      "type": "string",
+      "description": "Official registered name of the organization",
+      "required": true,
+      "min_value": null,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "phone_number",
+      "type": "string",
+      "description": "Primary contact phone number",
+      "required": true,
+      "min_value": null,
+      "max_value": null,
+      "pattern": "^\\([0-9]{3}\\) [0-9]{3}-[0-9]{4}$"
+    },
+    {
+      "name": "website_url",
+      "type": "string",
+      "description": "Organization's website address",
+      "required": true,
+      "min_value": null,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "return_type",
+      "type": "string",
+      "description": "Type of IRS return filed (990-PF for private foundations)",
+      "required": true,
+      "min_value": null,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "amended_return",
+      "type": "string",
+      "description": "Indicates if this is an amended return (Yes/No)",
+      "required": true,
+      "min_value": null,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "group_exemption_number",
+      "type": "string",
+      "description": "IRS group exemption number, if applicable",
+      "required": true,
+      "min_value": null,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "subsection_code",
+      "type": "string",
+      "description": "IRS subsection code (typically 501(c)(3) for foundations)",
+      "required": true,
+      "min_value": null,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "ruling_date",
+      "type": "string",
+      "description": "Date of IRS ruling or determination letter",
+      "required": true,
+      "min_value": null,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "accounting_method",
+      "type": "string",
+      "description": "Accounting method used (Cash, Accrual, or Other)",
+      "required": true,
+      "min_value": null,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "organization_type",
+      "type": "string",
+      "description": "Legal structure (corporation, trust, association, etc.)",
+      "required": true,
+      "min_value": null,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "year_of_formation",
+      "type": "string",
+      "description": "Year the organization was established",
+      "required": true,
+      "min_value": null,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "incorporation_state",
+      "type": "string",
+      "description": "State where the organization was incorporated",
+      "required": true,
+      "min_value": null,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "total_revenue",
+      "type": "float",
+      "description": "Sum of all revenue sources for the year",
+      "required": true,
+      "min_value": 0,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "contributions_gifts_grants",
+      "type": "float",
+      "description": "Revenue from donations, contributions, and grants",
+      "required": true,
+      "min_value": 0,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "program_service_revenue",
+      "type": "float",
+      "description": "Revenue generated from program services",
+      "required": true,
+      "min_value": 0,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "membership_dues",
+      "type": "float",
+      "description": "Revenue from membership dues and assessments",
+      "required": true,
+      "min_value": 0,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "investment_income",
+      "type": "float",
+      "description": "Income from interest, dividends, and other investments",
+      "required": true,
+      "min_value": 0,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "gains_losses_sales_assets",
+      "type": "float",
+      "description": "Net gains or losses from sale of investments and assets",
+      "required": true,
+      "min_value": null,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "rental_income",
+      "type": "float",
+      "description": "Income from rental of real estate or equipment",
+      "required": true,
+      "min_value": 0,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "related_organizations_revenue",
+      "type": "float",
+      "description": "Revenue received from related organizations",
+      "required": true,
+      "min_value": 0,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "gaming_revenue",
+      "type": "float",
+      "description": "Revenue from gaming and gambling activities",
+      "required": true,
+      "min_value": 0,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "other_revenue",
+      "type": "float",
+      "description": "All other revenue not categorized elsewhere",
+      "required": true,
+      "min_value": 0,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "government_grants",
+      "type": "float",
+      "description": "Revenue from federal, state, and local government grants",
+      "required": true,
+      "min_value": 0,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "foreign_contributions",
+      "type": "float",
+      "description": "Revenue from foreign sources and contributors",
+      "required": true,
+      "min_value": 0,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "total_expenses",
+      "type": "float",
+      "description": "Sum of all organizational expenses for the year",
+      "required": true,
+      "min_value": 0,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "program_services_expenses",
+      "type": "float",
+      "description": "Direct expenses for charitable program activities",
+      "required": true,
+      "min_value": 0,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "management_general_expenses",
+      "type": "float",
+      "description": "Administrative and general operating expenses",
+      "required": true,
+      "min_value": 0,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "fundraising_expenses",
+      "type": "float",
+      "description": "Expenses related to fundraising activities",
+      "required": true,
+      "min_value": 0,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "grants_us_organizations",
+      "type": "float",
+      "description": "Grants and assistance provided to domestic organizations",
+      "required": true,
+      "min_value": 0,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "grants_us_individuals",
+      "type": "float",
+      "description": "Grants and assistance provided to domestic individuals",
+      "required": true,
+      "min_value": 0,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "grants_foreign_organizations",
+      "type": "float",
+      "description": "Grants and assistance provided to foreign organizations",
+      "required": true,
+      "min_value": 0,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "grants_foreign_individuals",
+      "type": "float",
+      "description": "Grants and assistance provided to foreign individuals",
+      "required": true,
+      "min_value": 0,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "compensation_officers",
+      "type": "float",
+      "description": "Total compensation paid to officers and key employees",
+      "required": true,
+      "min_value": 0,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "compensation_other_staff",
+      "type": "float",
+      "description": "Compensation paid to other employees",
+      "required": true,
+      "min_value": 0,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "payroll_taxes_benefits",
+      "type": "float",
+      "description": "Payroll taxes, pension plans, and employee benefits",
+      "required": true,
+      "min_value": 0,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "professional_fees",
+      "type": "float",
+      "description": "Legal, accounting, and other professional service fees",
+      "required": true,
+      "min_value": 0,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "office_occupancy_costs",
+      "type": "float",
+      "description": "Rent, utilities, and facility-related expenses",
+      "required": true,
+      "min_value": 0,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "information_technology_costs",
+      "type": "float",
+      "description": "IT equipment, software, and technology expenses",
+      "required": true,
+      "min_value": 0,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "travel_conference_expenses",
+      "type": "float",
+      "description": "Travel, conferences, conventions, and meetings",
+      "required": true,
+      "min_value": 0,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "depreciation_amortization",
+      "type": "float",
+      "description": "Depreciation of equipment and amortization of intangibles",
+      "required": true,
+      "min_value": 0,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "insurance",
+      "type": "float",
+      "description": "Insurance premiums and related costs",
+      "required": true,
+      "min_value": 0,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "officers_list",
+      "type": "array_string",
+      "description": "JSON array of officers, directors, trustees, and key employees with their details",
+      "required": true,
+      "min_value": null,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "governing_body_size",
+      "type": "integer",
+      "description": "Total number of voting members on the governing body",
+      "required": true,
+      "min_value": 0,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "independent_members",
+      "type": "integer",
+      "description": "Number of independent voting members",
+      "required": true,
+      "min_value": 0,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "financial_statements_reviewed",
+      "type": "string",
+      "description": "Whether financial statements were reviewed or audited",
+      "required": true,
+      "min_value": null,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "form_990_provided_to_governing_body",
+      "type": "string",
+      "description": "Whether Form 990 was provided to governing body before filing",
+      "required": true,
+      "min_value": null,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "conflict_of_interest_policy",
+      "type": "string",
+      "description": "Whether organization has a conflict of interest policy",
+      "required": true,
+      "min_value": null,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "whistleblower_policy",
+      "type": "string",
+      "description": "Whether organization has a whistleblower policy",
+      "required": true,
+      "min_value": null,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "document_retention_policy",
+      "type": "string",
+      "description": "Whether organization has a document retention and destruction policy",
+      "required": true,
+      "min_value": null,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "ceo_compensation_review_process",
+      "type": "string",
+      "description": "Process used to determine compensation of organization's top management",
+      "required": true,
+      "min_value": null,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "public_disclosure_practices",
+      "type": "string",
+      "description": "How organization makes its governing documents and annual returns available to the public",
+      "required": true,
+      "min_value": null,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "program_accomplishments_list",
+      "type": "array_string",
+      "description": "JSON array of program service accomplishments with descriptions and financial details",
+      "required": true,
+      "min_value": null,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "total_fundraising_event_revenue",
+      "type": "float",
+      "description": "Total revenue from all fundraising events",
+      "required": true,
+      "min_value": 0,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "total_fundraising_event_expenses",
+      "type": "float",
+      "description": "Total direct expenses for all fundraising events",
+      "required": true,
+      "min_value": 0,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "professional_fundraiser_fees",
+      "type": "float",
+      "description": "Fees paid to professional fundraising services",
+      "required": true,
+      "min_value": 0,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "number_of_employees",
+      "type": "integer",
+      "description": "Total number of employees during the year",
+      "required": true,
+      "min_value": 0,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "number_of_volunteers",
+      "type": "integer",
+      "description": "Estimate of volunteers who provided services",
+      "required": true,
+      "min_value": 0,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "occupancy_costs",
+      "type": "float",
+      "description": "Total costs for office space and facilities",
+      "required": true,
+      "min_value": 0,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "fundraising_method_descriptions",
+      "type": "string",
+      "description": "Description of methods used for fundraising",
+      "required": true,
+      "min_value": null,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "joint_ventures_disregarded_entities",
+      "type": "string",
+      "description": "Information about joint ventures and disregarded entities",
+      "required": true,
+      "min_value": null,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "base_compensation",
+      "type": "float",
+      "description": "Base salary or wages paid to key personnel",
+      "required": true,
+      "min_value": 0,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "bonus",
+      "type": "float",
+      "description": "Bonus and incentive compensation paid",
+      "required": true,
+      "min_value": 0,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "incentive",
+      "type": "float",
+      "description": "Other incentive compensation",
+      "required": true,
+      "min_value": 0,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "other_compensation",
+      "type": "float",
+      "description": "Other forms of compensation",
+      "required": true,
+      "min_value": 0,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "non_fixed_compensation",
+      "type": "string",
+      "description": "Whether compensation arrangement is non-fixed",
+      "required": true,
+      "min_value": null,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "first_class_travel",
+      "type": "string",
+      "description": "Whether first-class or charter travel was provided",
+      "required": true,
+      "min_value": null,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "housing_allowance",
+      "type": "string",
+      "description": "Whether housing allowance or residence was provided",
+      "required": true,
+      "min_value": null,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "expense_account_usage",
+      "type": "string",
+      "description": "Whether payments for business use of personal residence were made",
+      "required": true,
+      "min_value": null,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "supplemental_retirement",
+      "type": "string",
+      "description": "Whether supplemental nonqualified retirement plan was provided",
+      "required": true,
+      "min_value": null,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "lobbying_expenditures_direct",
+      "type": "float",
+      "description": "Amount spent on direct lobbying activities",
+      "required": true,
+      "min_value": 0,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "lobbying_expenditures_grassroots",
+      "type": "float",
+      "description": "Amount spent on grassroots lobbying activities",
+      "required": true,
+      "min_value": 0,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "election_501h_status",
+      "type": "string",
+      "description": "Whether the organization made a Section 501(h) election",
+      "required": true,
+      "min_value": null,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "political_campaign_expenditures",
+      "type": "float",
+      "description": "Amount spent on political campaign activities",
+      "required": true,
+      "min_value": 0,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "related_organizations_affiliates",
+      "type": "string",
+      "description": "Information about related organizations involved in political activities",
+      "required": true,
+      "min_value": null,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "investment_types",
+      "type": "string",
+      "description": "Description of types of investments held",
+      "required": true,
+      "min_value": null,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "donor_restricted_endowment_values",
+      "type": "float",
+      "description": "Value of permanently restricted endowment funds",
+      "required": true,
+      "min_value": 0,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "net_appreciation_depreciation",
+      "type": "float",
+      "description": "Net appreciation or depreciation in fair value of investments",
+      "required": true,
+      "min_value": null,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "related_organization_transactions",
+      "type": "string",
+      "description": "Information about transactions with related organizations",
+      "required": true,
+      "min_value": null,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "loans_to_from_related_parties",
+      "type": "string",
+      "description": "Information about loans to or from related parties",
+      "required": true,
+      "min_value": null,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "penalties_excise_taxes_reported",
+      "type": "string",
+      "description": "Whether the organization reported any penalties or excise taxes",
+      "required": true,
+      "min_value": null,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "unrelated_business_income_disclosure",
+      "type": "string",
+      "description": "Whether the organization had unrelated business income",
+      "required": true,
+      "min_value": null,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "foreign_bank_account_reporting",
+      "type": "string",
+      "description": "Whether the organization had foreign bank accounts or assets",
+      "required": true,
+      "min_value": null,
+      "max_value": null,
+      "pattern": null
+    },
+    {
+      "name": "schedule_o_narrative_explanations",
+      "type": "string",
+      "description": "Additional narrative explanations from Schedule O",
+      "required": true,
+      "min_value": null,
+      "max_value": null,
+      "pattern": null
+    }
+  ],
+  "created_at": "2025-11-07T23:45:00.000000",
+  "updated_at": "2025-11-07T23:45:00.000000",
+  "tema": "IRS_FORM_990PF",
+  "is_global": true
+}

backend/pyproject.toml

@@ -28,8 +28,14 @@ dependencies = [
     # LandingAI Document AI
     "landingai-ade>=0.2.1",
     "redis-om>=0.3.5",
-    "pydantic-ai-slim[google,openai]>=1.11.1",
+    "pydantic-ai-slim[google,openai,mcp]>=1.11.1",
+    "tavily-python>=0.5.0",
 ]
 [project.scripts]
 dev = "uvicorn app.main:app --host 0.0.0.0 --port 8000 --reload"
 start = "uvicorn app.main:app --host 0.0.0.0 --port 8000"
+
+[dependency-groups]
+dev = [
+    "ruff>=0.14.4",
+]

backend/uv.lock

@@ -1,5 +1,5 @@
 version = 1
-revision = 3
+revision = 2
 requires-python = ">=3.12"
 resolution-markers = [
     "python_full_version >= '3.14'",
@@ -30,6 +30,15 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/6f/12/e5e0282d673bb9746bacfb6e2dba8719989d3660cdb2ea79aee9a9651afb/anyio-4.10.0-py3-none-any.whl", hash = "sha256:60e474ac86736bbfd6f210f7a61218939c318f43f9972497381f1c5e930ed3d1", size = 107213, upload-time = "2025-08-04T08:54:24.882Z" },
 ]
 
+[[package]]
+name = "attrs"
+version = "25.4.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/6b/5c/685e6633917e101e5dcb62b9dd76946cbb57c26e133bae9e0cd36033c0a9/attrs-25.4.0.tar.gz", hash = "sha256:16d5969b87f0859ef33a48b35d55ac1be6e42ae49d5e853b597db70c35c57e11", size = 934251, upload-time = "2025-10-06T13:54:44.725Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/3a/2a/7cc015f5b9f5db42b7d48157e23356022889fc354a2813c15934b7cb5c0e/attrs-25.4.0-py3-none-any.whl", hash = "sha256:adcf7e2a1fb3b36ac48d97835bb6d8ade15b8dcce26aba8bf1d14847b57a3373", size = 67615, upload-time = "2025-10-06T13:54:43.17Z" },
+]
+
 [[package]]
 name = "azure-core"
 version = "1.35.0"
@@ -74,18 +83,24 @@ dependencies = [
     { name = "openai" },
     { name = "pdf2image" },
     { name = "pillow" },
-    { name = "pydantic-ai-slim", extra = ["google", "openai"] },
+    { name = "pydantic-ai-slim", extra = ["google", "mcp", "openai"] },
     { name = "pydantic-settings" },
     { name = "pypdf" },
     { name = "python-dotenv" },
     { name = "python-multipart" },
     { name = "qdrant-client" },
     { name = "redis-om" },
+    { name = "tavily-python" },
     { name = "tiktoken" },
     { name = "uvicorn", extra = ["standard"] },
     { name = "websockets" },
 ]
 
+[package.dev-dependencies]
+dev = [
+    { name = "ruff" },
+]
+
 [package.metadata]
 requires-dist = [
     { name = "azure-storage-blob", specifier = ">=12.26.0" },
@@ -98,18 +113,22 @@ requires-dist = [
     { name = "openai", specifier = ">=1.59.6" },
     { name = "pdf2image", specifier = ">=1.17.0" },
     { name = "pillow", specifier = ">=11.0.0" },
-    { name = "pydantic-ai-slim", extras = ["google", "openai"], specifier = ">=1.11.1" },
+    { name = "pydantic-ai-slim", extras = ["google", "openai", "mcp"], specifier = ">=1.11.1" },
     { name = "pydantic-settings", specifier = ">=2.10.1" },
     { name = "pypdf", specifier = ">=5.1.0" },
     { name = "python-dotenv", specifier = ">=1.1.1" },
     { name = "python-multipart", specifier = ">=0.0.20" },
     { name = "qdrant-client", specifier = ">=1.15.1" },
     { name = "redis-om", specifier = ">=0.3.5" },
+    { name = "tavily-python", specifier = ">=0.5.0" },
     { name = "tiktoken", specifier = ">=0.8.0" },
     { name = "uvicorn", extras = ["standard"], specifier = ">=0.35.0" },
     { name = "websockets", specifier = ">=14.1" },
 ]
 
+[package.metadata.requires-dev]
+dev = [{ name = "ruff", specifier = ">=0.14.4" }]
+
 [[package]]
 name = "cachetools"
 version = "6.2.1"
@@ -785,6 +804,15 @@ http2 = [
     { name = "h2" },
 ]
 
+[[package]]
+name = "httpx-sse"
+version = "0.4.3"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/0f/4c/751061ffa58615a32c31b2d82e8482be8dd4a89154f003147acee90f2be9/httpx_sse-0.4.3.tar.gz", hash = "sha256:9b1ed0127459a66014aec3c56bebd93da3c1bc8bb6618c8082039a44889a755d", size = 15943, upload-time = "2025-10-10T21:48:22.271Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/d2/fd/6668e5aec43ab844de6fc74927e155a3b37bf40d7c3790e49fc0406b6578/httpx_sse-0.4.3-py3-none-any.whl", hash = "sha256:0ac1c9fe3c0afad2e0ebb25a934a59f4c7823b60792691f779fad2c5568830fc", size = 8960, upload-time = "2025-10-10T21:48:21.158Z" },
+]
+
 [[package]]
 name = "hyperframe"
 version = "6.1.0"
@@ -913,6 +941,33 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/71/92/5e77f98553e9e75130c78900d000368476aed74276eb8ae8796f65f00918/jsonpointer-3.0.0-py2.py3-none-any.whl", hash = "sha256:13e088adc14fca8b6aa8177c044e12701e6ad4b28ff10e65f2267a90109c9942", size = 7595, upload-time = "2024-06-10T19:24:40.698Z" },
 ]
 
+[[package]]
+name = "jsonschema"
+version = "4.25.1"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "attrs" },
+    { name = "jsonschema-specifications" },
+    { name = "referencing" },
+    { name = "rpds-py" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/74/69/f7185de793a29082a9f3c7728268ffb31cb5095131a9c139a74078e27336/jsonschema-4.25.1.tar.gz", hash = "sha256:e4a9655ce0da0c0b67a085847e00a3a51449e1157f4f75e9fb5aa545e122eb85", size = 357342, upload-time = "2025-08-18T17:03:50.038Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/bf/9c/8c95d856233c1f82500c2450b8c68576b4cf1c871db3afac5c34ff84e6fd/jsonschema-4.25.1-py3-none-any.whl", hash = "sha256:3fba0169e345c7175110351d456342c364814cfcf3b964ba4587f22915230a63", size = 90040, upload-time = "2025-08-18T17:03:48.373Z" },
+]
+
+[[package]]
+name = "jsonschema-specifications"
+version = "2025.9.1"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "referencing" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/19/74/a633ee74eb36c44aa6d1095e7cc5569bebf04342ee146178e2d36600708b/jsonschema_specifications-2025.9.1.tar.gz", hash = "sha256:b540987f239e745613c7a9176f3edb72b832a4ac465cf02712288397832b5e8d", size = 32855, upload-time = "2025-09-08T01:34:59.186Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/41/45/1a4ed80516f02155c51f51e8cedb3c1902296743db0bbc66608a0db2814f/jsonschema_specifications-2025.9.1-py3-none-any.whl", hash = "sha256:98802fee3a11ee76ecaca44429fda8a41bff98b00a0f2838151b113f210cc6fe", size = 18437, upload-time = "2025-09-08T01:34:57.871Z" },
+]
+
 [[package]]
 name = "landingai-ade"
 version = "0.20.3"
@@ -1057,6 +1112,29 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/58/c7/b06a83df678fca882c24fb498e628e0406bdb95ffdfa7ae43ecc0a714d52/logfire_api-4.14.2-py3-none-any.whl", hash = "sha256:aa4af2ecb007c3e0095e25ba4526fd8c0e2c0be2ceceac71ca651c4ad86dc713", size = 95021, upload-time = "2025-10-24T20:14:36.161Z" },
 ]
 
+[[package]]
+name = "mcp"
+version = "1.21.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "anyio" },
+    { name = "httpx" },
+    { name = "httpx-sse" },
+    { name = "jsonschema" },
+    { name = "pydantic" },
+    { name = "pydantic-settings" },
+    { name = "pyjwt", extra = ["crypto"] },
+    { name = "python-multipart" },
+    { name = "pywin32", marker = "sys_platform == 'win32'" },
+    { name = "sse-starlette" },
+    { name = "starlette" },
+    { name = "uvicorn", marker = "sys_platform != 'emscripten'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/33/54/dd2330ef4611c27ae59124820863c34e1d3edb1133c58e6375e2d938c9c5/mcp-1.21.0.tar.gz", hash = "sha256:bab0a38e8f8c48080d787233343f8d301b0e1e95846ae7dead251b2421d99855", size = 452697, upload-time = "2025-11-06T23:19:58.432Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/39/47/850b6edc96c03bd44b00de9a0ca3c1cc71e0ba1cd5822955bc9e4eb3fad3/mcp-1.21.0-py3-none-any.whl", hash = "sha256:598619e53eb0b7a6513db38c426b28a4bdf57496fed04332100d2c56acade98b", size = 173672, upload-time = "2025-11-06T23:19:56.508Z" },
+]
+
 [[package]]
 name = "more-itertools"
 version = "10.8.0"
@@ -1447,6 +1525,9 @@ wheels = [
 google = [
     { name = "google-genai" },
 ]
+mcp = [
+    { name = "mcp" },
+]
 openai = [
     { name = "openai" },
 ]
@@ -1531,6 +1612,11 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/61/ad/689f02752eeec26aed679477e80e632ef1b682313be70793d798c1d5fc8f/PyJWT-2.10.1-py3-none-any.whl", hash = "sha256:dcdd193e30abefd5debf142f9adfcdd2b58004e644f25406ffaebd50bd98dacb", size = 22997, upload-time = "2024-11-28T03:43:27.893Z" },
 ]
 
+[package.optional-dependencies]
+crypto = [
+    { name = "cryptography" },
+]
+
 [[package]]
 name = "pypdf"
 version = "6.1.3"
@@ -1672,6 +1758,20 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/e4/60/2cc6753c2c36a2a5dded8c380c6cad67a26c5878cd7aad56de2eee1d63c8/redis_om-0.3.5-py3-none-any.whl", hash = "sha256:99ab40f696028ce47c5e2eb5118a1ffc1fd193005428df89c8cf77ad35a0177a", size = 86634, upload-time = "2025-04-04T12:54:50.07Z" },
 ]
 
+[[package]]
+name = "referencing"
+version = "0.37.0"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "attrs" },
+    { name = "rpds-py" },
+    { name = "typing-extensions", marker = "python_full_version < '3.13'" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/22/f5/df4e9027acead3ecc63e50fe1e36aca1523e1719559c499951bb4b53188f/referencing-0.37.0.tar.gz", hash = "sha256:44aefc3142c5b842538163acb373e24cce6632bd54bdb01b21ad5863489f50d8", size = 78036, upload-time = "2025-10-13T15:30:48.871Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/2c/58/ca301544e1fa93ed4f80d724bf5b194f6e4b945841c5bfd555878eea9fcb/referencing-0.37.0-py3-none-any.whl", hash = "sha256:381329a9f99628c9069361716891d34ad94af76e461dcb0335825aecc7692231", size = 26766, upload-time = "2025-10-13T15:30:47.625Z" },
+]
+
 [[package]]
 name = "regex"
 version = "2025.11.3"
@@ -1777,6 +1877,87 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/3f/51/d4db610ef29373b879047326cbf6fa98b6c1969d6f6dc423279de2b1be2c/requests_toolbelt-1.0.0-py2.py3-none-any.whl", hash = "sha256:cccfdd665f0a24fcf4726e690f65639d272bb0637b9b92dfd91a5568ccf6bd06", size = 54481, upload-time = "2023-05-01T04:11:28.427Z" },
 ]
 
+[[package]]
+name = "rpds-py"
+version = "0.28.0"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/48/dc/95f074d43452b3ef5d06276696ece4b3b5d696e7c9ad7173c54b1390cd70/rpds_py-0.28.0.tar.gz", hash = "sha256:abd4df20485a0983e2ca334a216249b6186d6e3c1627e106651943dbdb791aea", size = 27419, upload-time = "2025-10-22T22:24:29.327Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/b8/5c/6c3936495003875fe7b14f90ea812841a08fca50ab26bd840e924097d9c8/rpds_py-0.28.0-cp312-cp312-macosx_10_12_x86_64.whl", hash = "sha256:6b4f28583a4f247ff60cd7bdda83db8c3f5b05a7a82ff20dd4b078571747708f", size = 366439, upload-time = "2025-10-22T22:22:04.525Z" },
+    { url = "https://files.pythonhosted.org/packages/56/f9/a0f1ca194c50aa29895b442771f036a25b6c41a35e4f35b1a0ea713bedae/rpds_py-0.28.0-cp312-cp312-macosx_11_0_arm64.whl", hash = "sha256:d678e91b610c29c4b3d52a2c148b641df2b4676ffe47c59f6388d58b99cdc424", size = 348170, upload-time = "2025-10-22T22:22:06.397Z" },
+    { url = "https://files.pythonhosted.org/packages/18/ea/42d243d3a586beb72c77fa5def0487daf827210069a95f36328e869599ea/rpds_py-0.28.0-cp312-cp312-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:e819e0e37a44a78e1383bf1970076e2ccc4dc8c2bbaa2f9bd1dc987e9afff628", size = 378838, upload-time = "2025-10-22T22:22:07.932Z" },
+    { url = "https://files.pythonhosted.org/packages/e7/78/3de32e18a94791af8f33601402d9d4f39613136398658412a4e0b3047327/rpds_py-0.28.0-cp312-cp312-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:5ee514e0f0523db5d3fb171f397c54875dbbd69760a414dccf9d4d7ad628b5bd", size = 393299, upload-time = "2025-10-22T22:22:09.435Z" },
+    { url = "https://files.pythonhosted.org/packages/13/7e/4bdb435afb18acea2eb8a25ad56b956f28de7c59f8a1d32827effa0d4514/rpds_py-0.28.0-cp312-cp312-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:5f3fa06d27fdcee47f07a39e02862da0100cb4982508f5ead53ec533cd5fe55e", size = 518000, upload-time = "2025-10-22T22:22:11.326Z" },
+    { url = "https://files.pythonhosted.org/packages/31/d0/5f52a656875cdc60498ab035a7a0ac8f399890cc1ee73ebd567bac4e39ae/rpds_py-0.28.0-cp312-cp312-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:46959ef2e64f9e4a41fc89aa20dbca2b85531f9a72c21099a3360f35d10b0d5a", size = 408746, upload-time = "2025-10-22T22:22:13.143Z" },
+    { url = "https://files.pythonhosted.org/packages/3e/cd/49ce51767b879cde77e7ad9fae164ea15dce3616fe591d9ea1df51152706/rpds_py-0.28.0-cp312-cp312-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:8455933b4bcd6e83fde3fefc987a023389c4b13f9a58c8d23e4b3f6d13f78c84", size = 386379, upload-time = "2025-10-22T22:22:14.602Z" },
+    { url = "https://files.pythonhosted.org/packages/6a/99/e4e1e1ee93a98f72fc450e36c0e4d99c35370220e815288e3ecd2ec36a2a/rpds_py-0.28.0-cp312-cp312-manylinux_2_31_riscv64.whl", hash = "sha256:ad50614a02c8c2962feebe6012b52f9802deec4263946cddea37aaf28dd25a66", size = 401280, upload-time = "2025-10-22T22:22:16.063Z" },
+    { url = "https://files.pythonhosted.org/packages/61/35/e0c6a57488392a8b319d2200d03dad2b29c0db9996f5662c3b02d0b86c02/rpds_py-0.28.0-cp312-cp312-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:e5deca01b271492553fdb6c7fd974659dce736a15bae5dad7ab8b93555bceb28", size = 412365, upload-time = "2025-10-22T22:22:17.504Z" },
+    { url = "https://files.pythonhosted.org/packages/ff/6a/841337980ea253ec797eb084665436007a1aad0faac1ba097fb906c5f69c/rpds_py-0.28.0-cp312-cp312-musllinux_1_2_aarch64.whl", hash = "sha256:735f8495a13159ce6a0d533f01e8674cec0c57038c920495f87dcb20b3ddb48a", size = 559573, upload-time = "2025-10-22T22:22:19.108Z" },
+    { url = "https://files.pythonhosted.org/packages/e7/5e/64826ec58afd4c489731f8b00729c5f6afdb86f1df1df60bfede55d650bb/rpds_py-0.28.0-cp312-cp312-musllinux_1_2_i686.whl", hash = "sha256:961ca621ff10d198bbe6ba4957decca61aa2a0c56695384c1d6b79bf61436df5", size = 583973, upload-time = "2025-10-22T22:22:20.768Z" },
+    { url = "https://files.pythonhosted.org/packages/b6/ee/44d024b4843f8386a4eeaa4c171b3d31d55f7177c415545fd1a24c249b5d/rpds_py-0.28.0-cp312-cp312-musllinux_1_2_x86_64.whl", hash = "sha256:2374e16cc9131022e7d9a8f8d65d261d9ba55048c78f3b6e017971a4f5e6353c", size = 553800, upload-time = "2025-10-22T22:22:22.25Z" },
+    { url = "https://files.pythonhosted.org/packages/7d/89/33e675dccff11a06d4d85dbb4d1865f878d5020cbb69b2c1e7b2d3f82562/rpds_py-0.28.0-cp312-cp312-win32.whl", hash = "sha256:d15431e334fba488b081d47f30f091e5d03c18527c325386091f31718952fe08", size = 216954, upload-time = "2025-10-22T22:22:24.105Z" },
+    { url = "https://files.pythonhosted.org/packages/af/36/45f6ebb3210887e8ee6dbf1bc710ae8400bb417ce165aaf3024b8360d999/rpds_py-0.28.0-cp312-cp312-win_amd64.whl", hash = "sha256:a410542d61fc54710f750d3764380b53bf09e8c4edbf2f9141a82aa774a04f7c", size = 227844, upload-time = "2025-10-22T22:22:25.551Z" },
+    { url = "https://files.pythonhosted.org/packages/57/91/f3fb250d7e73de71080f9a221d19bd6a1c1eb0d12a1ea26513f6c1052ad6/rpds_py-0.28.0-cp312-cp312-win_arm64.whl", hash = "sha256:1f0cfd1c69e2d14f8c892b893997fa9a60d890a0c8a603e88dca4955f26d1edd", size = 217624, upload-time = "2025-10-22T22:22:26.914Z" },
+    { url = "https://files.pythonhosted.org/packages/d3/03/ce566d92611dfac0085c2f4b048cd53ed7c274a5c05974b882a908d540a2/rpds_py-0.28.0-cp313-cp313-macosx_10_12_x86_64.whl", hash = "sha256:e9e184408a0297086f880556b6168fa927d677716f83d3472ea333b42171ee3b", size = 366235, upload-time = "2025-10-22T22:22:28.397Z" },
+    { url = "https://files.pythonhosted.org/packages/00/34/1c61da1b25592b86fd285bd7bd8422f4c9d748a7373b46126f9ae792a004/rpds_py-0.28.0-cp313-cp313-macosx_11_0_arm64.whl", hash = "sha256:edd267266a9b0448f33dc465a97cfc5d467594b600fe28e7fa2f36450e03053a", size = 348241, upload-time = "2025-10-22T22:22:30.171Z" },
+    { url = "https://files.pythonhosted.org/packages/fc/00/ed1e28616848c61c493a067779633ebf4b569eccaacf9ccbdc0e7cba2b9d/rpds_py-0.28.0-cp313-cp313-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:85beb8b3f45e4e32f6802fb6cd6b17f615ef6c6a52f265371fb916fae02814aa", size = 378079, upload-time = "2025-10-22T22:22:31.644Z" },
+    { url = "https://files.pythonhosted.org/packages/11/b2/ccb30333a16a470091b6e50289adb4d3ec656fd9951ba8c5e3aaa0746a67/rpds_py-0.28.0-cp313-cp313-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:d2412be8d00a1b895f8ad827cc2116455196e20ed994bb704bf138fe91a42724", size = 393151, upload-time = "2025-10-22T22:22:33.453Z" },
+    { url = "https://files.pythonhosted.org/packages/8c/d0/73e2217c3ee486d555cb84920597480627d8c0240ff3062005c6cc47773e/rpds_py-0.28.0-cp313-cp313-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:cf128350d384b777da0e68796afdcebc2e9f63f0e9f242217754e647f6d32491", size = 517520, upload-time = "2025-10-22T22:22:34.949Z" },
+    { url = "https://files.pythonhosted.org/packages/c4/91/23efe81c700427d0841a4ae7ea23e305654381831e6029499fe80be8a071/rpds_py-0.28.0-cp313-cp313-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:a2036d09b363aa36695d1cc1a97b36865597f4478470b0697b5ee9403f4fe399", size = 408699, upload-time = "2025-10-22T22:22:36.584Z" },
+    { url = "https://files.pythonhosted.org/packages/ca/ee/a324d3198da151820a326c1f988caaa4f37fc27955148a76fff7a2d787a9/rpds_py-0.28.0-cp313-cp313-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:b8e1e9be4fa6305a16be628959188e4fd5cd6f1b0e724d63c6d8b2a8adf74ea6", size = 385720, upload-time = "2025-10-22T22:22:38.014Z" },
+    { url = "https://files.pythonhosted.org/packages/19/ad/e68120dc05af8b7cab4a789fccd8cdcf0fe7e6581461038cc5c164cd97d2/rpds_py-0.28.0-cp313-cp313-manylinux_2_31_riscv64.whl", hash = "sha256:0a403460c9dd91a7f23fc3188de6d8977f1d9603a351d5db6cf20aaea95b538d", size = 401096, upload-time = "2025-10-22T22:22:39.869Z" },
+    { url = "https://files.pythonhosted.org/packages/99/90/c1e070620042459d60df6356b666bb1f62198a89d68881816a7ed121595a/rpds_py-0.28.0-cp313-cp313-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:d7366b6553cdc805abcc512b849a519167db8f5e5c3472010cd1228b224265cb", size = 411465, upload-time = "2025-10-22T22:22:41.395Z" },
+    { url = "https://files.pythonhosted.org/packages/68/61/7c195b30d57f1b8d5970f600efee72a4fad79ec829057972e13a0370fd24/rpds_py-0.28.0-cp313-cp313-musllinux_1_2_aarch64.whl", hash = "sha256:5b43c6a3726efd50f18d8120ec0551241c38785b68952d240c45ea553912ac41", size = 558832, upload-time = "2025-10-22T22:22:42.871Z" },
+    { url = "https://files.pythonhosted.org/packages/b0/3d/06f3a718864773f69941d4deccdf18e5e47dd298b4628062f004c10f3b34/rpds_py-0.28.0-cp313-cp313-musllinux_1_2_i686.whl", hash = "sha256:0cb7203c7bc69d7c1585ebb33a2e6074492d2fc21ad28a7b9d40457ac2a51ab7", size = 583230, upload-time = "2025-10-22T22:22:44.877Z" },
+    { url = "https://files.pythonhosted.org/packages/66/df/62fc783781a121e77fee9a21ead0a926f1b652280a33f5956a5e7833ed30/rpds_py-0.28.0-cp313-cp313-musllinux_1_2_x86_64.whl", hash = "sha256:7a52a5169c664dfb495882adc75c304ae1d50df552fbd68e100fdc719dee4ff9", size = 553268, upload-time = "2025-10-22T22:22:46.441Z" },
+    { url = "https://files.pythonhosted.org/packages/84/85/d34366e335140a4837902d3dea89b51f087bd6a63c993ebdff59e93ee61d/rpds_py-0.28.0-cp313-cp313-win32.whl", hash = "sha256:2e42456917b6687215b3e606ab46aa6bca040c77af7df9a08a6dcfe8a4d10ca5", size = 217100, upload-time = "2025-10-22T22:22:48.342Z" },
+    { url = "https://files.pythonhosted.org/packages/3c/1c/f25a3f3752ad7601476e3eff395fe075e0f7813fbb9862bd67c82440e880/rpds_py-0.28.0-cp313-cp313-win_amd64.whl", hash = "sha256:e0a0311caedc8069d68fc2bf4c9019b58a2d5ce3cd7cb656c845f1615b577e1e", size = 227759, upload-time = "2025-10-22T22:22:50.219Z" },
+    { url = "https://files.pythonhosted.org/packages/e0/d6/5f39b42b99615b5bc2f36ab90423ea404830bdfee1c706820943e9a645eb/rpds_py-0.28.0-cp313-cp313-win_arm64.whl", hash = "sha256:04c1b207ab8b581108801528d59ad80aa83bb170b35b0ddffb29c20e411acdc1", size = 217326, upload-time = "2025-10-22T22:22:51.647Z" },
+    { url = "https://files.pythonhosted.org/packages/5c/8b/0c69b72d1cee20a63db534be0df271effe715ef6c744fdf1ff23bb2b0b1c/rpds_py-0.28.0-cp313-cp313t-macosx_10_12_x86_64.whl", hash = "sha256:f296ea3054e11fc58ad42e850e8b75c62d9a93a9f981ad04b2e5ae7d2186ff9c", size = 355736, upload-time = "2025-10-22T22:22:53.211Z" },
+    { url = "https://files.pythonhosted.org/packages/f7/6d/0c2ee773cfb55c31a8514d2cece856dd299170a49babd50dcffb15ddc749/rpds_py-0.28.0-cp313-cp313t-macosx_11_0_arm64.whl", hash = "sha256:5a7306c19b19005ad98468fcefeb7100b19c79fc23a5f24a12e06d91181193fa", size = 342677, upload-time = "2025-10-22T22:22:54.723Z" },
+    { url = "https://files.pythonhosted.org/packages/e2/1c/22513ab25a27ea205144414724743e305e8153e6abe81833b5e678650f5a/rpds_py-0.28.0-cp313-cp313t-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:e5d9b86aa501fed9862a443c5c3116f6ead8bc9296185f369277c42542bd646b", size = 371847, upload-time = "2025-10-22T22:22:56.295Z" },
+    { url = "https://files.pythonhosted.org/packages/60/07/68e6ccdb4b05115ffe61d31afc94adef1833d3a72f76c9632d4d90d67954/rpds_py-0.28.0-cp313-cp313t-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:e5bbc701eff140ba0e872691d573b3d5d30059ea26e5785acba9132d10c8c31d", size = 381800, upload-time = "2025-10-22T22:22:57.808Z" },
+    { url = "https://files.pythonhosted.org/packages/73/bf/6d6d15df80781d7f9f368e7c1a00caf764436518c4877fb28b029c4624af/rpds_py-0.28.0-cp313-cp313t-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:9a5690671cd672a45aa8616d7374fdf334a1b9c04a0cac3c854b1136e92374fe", size = 518827, upload-time = "2025-10-22T22:22:59.826Z" },
+    { url = "https://files.pythonhosted.org/packages/7b/d3/2decbb2976cc452cbf12a2b0aaac5f1b9dc5dd9d1f7e2509a3ee00421249/rpds_py-0.28.0-cp313-cp313t-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:9f1d92ecea4fa12f978a367c32a5375a1982834649cdb96539dcdc12e609ab1a", size = 399471, upload-time = "2025-10-22T22:23:01.968Z" },
+    { url = "https://files.pythonhosted.org/packages/b1/2c/f30892f9e54bd02e5faca3f6a26d6933c51055e67d54818af90abed9748e/rpds_py-0.28.0-cp313-cp313t-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:8d252db6b1a78d0a3928b6190156042d54c93660ce4d98290d7b16b5296fb7cc", size = 377578, upload-time = "2025-10-22T22:23:03.52Z" },
+    { url = "https://files.pythonhosted.org/packages/f0/5d/3bce97e5534157318f29ac06bf2d279dae2674ec12f7cb9c12739cee64d8/rpds_py-0.28.0-cp313-cp313t-manylinux_2_31_riscv64.whl", hash = "sha256:d61b355c3275acb825f8777d6c4505f42b5007e357af500939d4a35b19177259", size = 390482, upload-time = "2025-10-22T22:23:05.391Z" },
+    { url = "https://files.pythonhosted.org/packages/e3/f0/886bd515ed457b5bd93b166175edb80a0b21a210c10e993392127f1e3931/rpds_py-0.28.0-cp313-cp313t-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:acbe5e8b1026c0c580d0321c8aae4b0a1e1676861d48d6e8c6586625055b606a", size = 402447, upload-time = "2025-10-22T22:23:06.93Z" },
+    { url = "https://files.pythonhosted.org/packages/42/b5/71e8777ac55e6af1f4f1c05b47542a1eaa6c33c1cf0d300dca6a1c6e159a/rpds_py-0.28.0-cp313-cp313t-musllinux_1_2_aarch64.whl", hash = "sha256:8aa23b6f0fc59b85b4c7d89ba2965af274346f738e8d9fc2455763602e62fd5f", size = 552385, upload-time = "2025-10-22T22:23:08.557Z" },
+    { url = "https://files.pythonhosted.org/packages/5d/cb/6ca2d70cbda5a8e36605e7788c4aa3bea7c17d71d213465a5a675079b98d/rpds_py-0.28.0-cp313-cp313t-musllinux_1_2_i686.whl", hash = "sha256:7b14b0c680286958817c22d76fcbca4800ddacef6f678f3a7c79a1fe7067fe37", size = 575642, upload-time = "2025-10-22T22:23:10.348Z" },
+    { url = "https://files.pythonhosted.org/packages/4a/d4/407ad9960ca7856d7b25c96dcbe019270b5ffdd83a561787bc682c797086/rpds_py-0.28.0-cp313-cp313t-musllinux_1_2_x86_64.whl", hash = "sha256:bcf1d210dfee61a6c86551d67ee1031899c0fdbae88b2d44a569995d43797712", size = 544507, upload-time = "2025-10-22T22:23:12.434Z" },
+    { url = "https://files.pythonhosted.org/packages/51/31/2f46fe0efcac23fbf5797c6b6b7e1c76f7d60773e525cb65fcbc582ee0f2/rpds_py-0.28.0-cp313-cp313t-win32.whl", hash = "sha256:3aa4dc0fdab4a7029ac63959a3ccf4ed605fee048ba67ce89ca3168da34a1342", size = 205376, upload-time = "2025-10-22T22:23:13.979Z" },
+    { url = "https://files.pythonhosted.org/packages/92/e4/15947bda33cbedfc134490a41841ab8870a72a867a03d4969d886f6594a2/rpds_py-0.28.0-cp313-cp313t-win_amd64.whl", hash = "sha256:7b7d9d83c942855e4fdcfa75d4f96f6b9e272d42fffcb72cd4bb2577db2e2907", size = 215907, upload-time = "2025-10-22T22:23:15.5Z" },
+    { url = "https://files.pythonhosted.org/packages/08/47/ffe8cd7a6a02833b10623bf765fbb57ce977e9a4318ca0e8cf97e9c3d2b3/rpds_py-0.28.0-cp314-cp314-macosx_10_12_x86_64.whl", hash = "sha256:dcdcb890b3ada98a03f9f2bb108489cdc7580176cb73b4f2d789e9a1dac1d472", size = 353830, upload-time = "2025-10-22T22:23:17.03Z" },
+    { url = "https://files.pythonhosted.org/packages/f9/9f/890f36cbd83a58491d0d91ae0db1702639edb33fb48eeb356f80ecc6b000/rpds_py-0.28.0-cp314-cp314-macosx_11_0_arm64.whl", hash = "sha256:f274f56a926ba2dc02976ca5b11c32855cbd5925534e57cfe1fda64e04d1add2", size = 341819, upload-time = "2025-10-22T22:23:18.57Z" },
+    { url = "https://files.pythonhosted.org/packages/09/e3/921eb109f682aa24fb76207698fbbcf9418738f35a40c21652c29053f23d/rpds_py-0.28.0-cp314-cp314-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:4fe0438ac4a29a520ea94c8c7f1754cdd8feb1bc490dfda1bfd990072363d527", size = 373127, upload-time = "2025-10-22T22:23:20.216Z" },
+    { url = "https://files.pythonhosted.org/packages/23/13/bce4384d9f8f4989f1a9599c71b7a2d877462e5fd7175e1f69b398f729f4/rpds_py-0.28.0-cp314-cp314-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:8a358a32dd3ae50e933347889b6af9a1bdf207ba5d1a3f34e1a38cd3540e6733", size = 382767, upload-time = "2025-10-22T22:23:21.787Z" },
+    { url = "https://files.pythonhosted.org/packages/23/e1/579512b2d89a77c64ccef5a0bc46a6ef7f72ae0cf03d4b26dcd52e57ee0a/rpds_py-0.28.0-cp314-cp314-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:e80848a71c78aa328fefaba9c244d588a342c8e03bda518447b624ea64d1ff56", size = 517585, upload-time = "2025-10-22T22:23:23.699Z" },
+    { url = "https://files.pythonhosted.org/packages/62/3c/ca704b8d324a2591b0b0adcfcaadf9c862375b11f2f667ac03c61b4fd0a6/rpds_py-0.28.0-cp314-cp314-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:f586db2e209d54fe177e58e0bc4946bea5fb0102f150b1b2f13de03e1f0976f8", size = 399828, upload-time = "2025-10-22T22:23:25.713Z" },
+    { url = "https://files.pythonhosted.org/packages/da/37/e84283b9e897e3adc46b4c88bb3f6ec92a43bd4d2f7ef5b13459963b2e9c/rpds_py-0.28.0-cp314-cp314-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:5ae8ee156d6b586e4292491e885d41483136ab994e719a13458055bec14cf370", size = 375509, upload-time = "2025-10-22T22:23:27.32Z" },
+    { url = "https://files.pythonhosted.org/packages/1a/c2/a980beab869d86258bf76ec42dec778ba98151f253a952b02fe36d72b29c/rpds_py-0.28.0-cp314-cp314-manylinux_2_31_riscv64.whl", hash = "sha256:a805e9b3973f7e27f7cab63a6b4f61d90f2e5557cff73b6e97cd5b8540276d3d", size = 392014, upload-time = "2025-10-22T22:23:29.332Z" },
+    { url = "https://files.pythonhosted.org/packages/da/b5/b1d3c5f9d3fa5aeef74265f9c64de3c34a0d6d5cd3c81c8b17d5c8f10ed4/rpds_py-0.28.0-cp314-cp314-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:5d3fd16b6dc89c73a4da0b4ac8b12a7ecc75b2864b95c9e5afed8003cb50a728", size = 402410, upload-time = "2025-10-22T22:23:31.14Z" },
+    { url = "https://files.pythonhosted.org/packages/74/ae/cab05ff08dfcc052afc73dcb38cbc765ffc86f94e966f3924cd17492293c/rpds_py-0.28.0-cp314-cp314-musllinux_1_2_aarch64.whl", hash = "sha256:6796079e5d24fdaba6d49bda28e2c47347e89834678f2bc2c1b4fc1489c0fb01", size = 553593, upload-time = "2025-10-22T22:23:32.834Z" },
+    { url = "https://files.pythonhosted.org/packages/70/80/50d5706ea2a9bfc9e9c5f401d91879e7c790c619969369800cde202da214/rpds_py-0.28.0-cp314-cp314-musllinux_1_2_i686.whl", hash = "sha256:76500820c2af232435cbe215e3324c75b950a027134e044423f59f5b9a1ba515", size = 576925, upload-time = "2025-10-22T22:23:34.47Z" },
+    { url = "https://files.pythonhosted.org/packages/ab/12/85a57d7a5855a3b188d024b099fd09c90db55d32a03626d0ed16352413ff/rpds_py-0.28.0-cp314-cp314-musllinux_1_2_x86_64.whl", hash = "sha256:bbdc5640900a7dbf9dd707fe6388972f5bbd883633eb68b76591044cfe346f7e", size = 542444, upload-time = "2025-10-22T22:23:36.093Z" },
+    { url = "https://files.pythonhosted.org/packages/6c/65/10643fb50179509150eb94d558e8837c57ca8b9adc04bd07b98e57b48f8c/rpds_py-0.28.0-cp314-cp314-win32.whl", hash = "sha256:adc8aa88486857d2b35d75f0640b949759f79dc105f50aa2c27816b2e0dd749f", size = 207968, upload-time = "2025-10-22T22:23:37.638Z" },
+    { url = "https://files.pythonhosted.org/packages/b4/84/0c11fe4d9aaea784ff4652499e365963222481ac647bcd0251c88af646eb/rpds_py-0.28.0-cp314-cp314-win_amd64.whl", hash = "sha256:66e6fa8e075b58946e76a78e69e1a124a21d9a48a5b4766d15ba5b06869d1fa1", size = 218876, upload-time = "2025-10-22T22:23:39.179Z" },
+    { url = "https://files.pythonhosted.org/packages/0f/e0/3ab3b86ded7bb18478392dc3e835f7b754cd446f62f3fc96f4fe2aca78f6/rpds_py-0.28.0-cp314-cp314-win_arm64.whl", hash = "sha256:a6fe887c2c5c59413353b7c0caff25d0e566623501ccfff88957fa438a69377d", size = 212506, upload-time = "2025-10-22T22:23:40.755Z" },
+    { url = "https://files.pythonhosted.org/packages/51/ec/d5681bb425226c3501eab50fc30e9d275de20c131869322c8a1729c7b61c/rpds_py-0.28.0-cp314-cp314t-macosx_10_12_x86_64.whl", hash = "sha256:7a69df082db13c7070f7b8b1f155fa9e687f1d6aefb7b0e3f7231653b79a067b", size = 355433, upload-time = "2025-10-22T22:23:42.259Z" },
+    { url = "https://files.pythonhosted.org/packages/be/ec/568c5e689e1cfb1ea8b875cffea3649260955f677fdd7ddc6176902d04cd/rpds_py-0.28.0-cp314-cp314t-macosx_11_0_arm64.whl", hash = "sha256:b1cde22f2c30ebb049a9e74c5374994157b9b70a16147d332f89c99c5960737a", size = 342601, upload-time = "2025-10-22T22:23:44.372Z" },
+    { url = "https://files.pythonhosted.org/packages/32/fe/51ada84d1d2a1d9d8f2c902cfddd0133b4a5eb543196ab5161d1c07ed2ad/rpds_py-0.28.0-cp314-cp314t-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:5338742f6ba7a51012ea470bd4dc600a8c713c0c72adaa0977a1b1f4327d6592", size = 372039, upload-time = "2025-10-22T22:23:46.025Z" },
+    { url = "https://files.pythonhosted.org/packages/07/c1/60144a2f2620abade1a78e0d91b298ac2d9b91bc08864493fa00451ef06e/rpds_py-0.28.0-cp314-cp314t-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:e1460ebde1bcf6d496d80b191d854adedcc619f84ff17dc1c6d550f58c9efbba", size = 382407, upload-time = "2025-10-22T22:23:48.098Z" },
+    { url = "https://files.pythonhosted.org/packages/45/ed/091a7bbdcf4038a60a461df50bc4c82a7ed6d5d5e27649aab61771c17585/rpds_py-0.28.0-cp314-cp314t-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:e3eb248f2feba84c692579257a043a7699e28a77d86c77b032c1d9fbb3f0219c", size = 518172, upload-time = "2025-10-22T22:23:50.16Z" },
+    { url = "https://files.pythonhosted.org/packages/54/dd/02cc90c2fd9c2ef8016fd7813bfacd1c3a1325633ec8f244c47b449fc868/rpds_py-0.28.0-cp314-cp314t-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:bd3bbba5def70b16cd1c1d7255666aad3b290fbf8d0fe7f9f91abafb73611a91", size = 399020, upload-time = "2025-10-22T22:23:51.81Z" },
+    { url = "https://files.pythonhosted.org/packages/ab/81/5d98cc0329bbb911ccecd0b9e19fbf7f3a5de8094b4cda5e71013b2dd77e/rpds_py-0.28.0-cp314-cp314t-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:3114f4db69ac5a1f32e7e4d1cbbe7c8f9cf8217f78e6e002cedf2d54c2a548ed", size = 377451, upload-time = "2025-10-22T22:23:53.711Z" },
+    { url = "https://files.pythonhosted.org/packages/b4/07/4d5bcd49e3dfed2d38e2dcb49ab6615f2ceb9f89f5a372c46dbdebb4e028/rpds_py-0.28.0-cp314-cp314t-manylinux_2_31_riscv64.whl", hash = "sha256:4b0cb8a906b1a0196b863d460c0222fb8ad0f34041568da5620f9799b83ccf0b", size = 390355, upload-time = "2025-10-22T22:23:55.299Z" },
+    { url = "https://files.pythonhosted.org/packages/3f/79/9f14ba9010fee74e4f40bf578735cfcbb91d2e642ffd1abe429bb0b96364/rpds_py-0.28.0-cp314-cp314t-manylinux_2_5_i686.manylinux1_i686.whl", hash = "sha256:cf681ac76a60b667106141e11a92a3330890257e6f559ca995fbb5265160b56e", size = 403146, upload-time = "2025-10-22T22:23:56.929Z" },
+    { url = "https://files.pythonhosted.org/packages/39/4c/f08283a82ac141331a83a40652830edd3a4a92c34e07e2bbe00baaea2f5f/rpds_py-0.28.0-cp314-cp314t-musllinux_1_2_aarch64.whl", hash = "sha256:1e8ee6413cfc677ce8898d9cde18cc3a60fc2ba756b0dec5b71eb6eb21c49fa1", size = 552656, upload-time = "2025-10-22T22:23:58.62Z" },
+    { url = "https://files.pythonhosted.org/packages/61/47/d922fc0666f0dd8e40c33990d055f4cc6ecff6f502c2d01569dbed830f9b/rpds_py-0.28.0-cp314-cp314t-musllinux_1_2_i686.whl", hash = "sha256:b3072b16904d0b5572a15eb9d31c1954e0d3227a585fc1351aa9878729099d6c", size = 576782, upload-time = "2025-10-22T22:24:00.312Z" },
+    { url = "https://files.pythonhosted.org/packages/d3/0c/5bafdd8ccf6aa9d3bfc630cfece457ff5b581af24f46a9f3590f790e3df2/rpds_py-0.28.0-cp314-cp314t-musllinux_1_2_x86_64.whl", hash = "sha256:b670c30fd87a6aec281c3c9896d3bae4b205fd75d79d06dc87c2503717e46092", size = 544671, upload-time = "2025-10-22T22:24:02.297Z" },
+    { url = "https://files.pythonhosted.org/packages/2c/37/dcc5d8397caa924988693519069d0beea077a866128719351a4ad95e82fc/rpds_py-0.28.0-cp314-cp314t-win32.whl", hash = "sha256:8014045a15b4d2b3476f0a287fcc93d4f823472d7d1308d47884ecac9e612be3", size = 205749, upload-time = "2025-10-22T22:24:03.848Z" },
+    { url = "https://files.pythonhosted.org/packages/d7/69/64d43b21a10d72b45939a28961216baeb721cc2a430f5f7c3bfa21659a53/rpds_py-0.28.0-cp314-cp314t-win_amd64.whl", hash = "sha256:7a4e59c90d9c27c561eb3160323634a9ff50b04e4f7820600a2beb0ac90db578", size = 216233, upload-time = "2025-10-22T22:24:05.471Z" },
+]
+
 [[package]]
 name = "rsa"
 version = "4.9.1"
@@ -1789,6 +1970,32 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/64/8d/0133e4eb4beed9e425d9a98ed6e081a55d195481b7632472be1af08d2f6b/rsa-4.9.1-py3-none-any.whl", hash = "sha256:68635866661c6836b8d39430f97a996acbd61bfa49406748ea243539fe239762", size = 34696, upload-time = "2025-04-16T09:51:17.142Z" },
 ]
 
+[[package]]
+name = "ruff"
+version = "0.14.4"
+source = { registry = "https://pypi.org/simple" }
+sdist = { url = "https://files.pythonhosted.org/packages/df/55/cccfca45157a2031dcbb5a462a67f7cf27f8b37d4b3b1cd7438f0f5c1df6/ruff-0.14.4.tar.gz", hash = "sha256:f459a49fe1085a749f15414ca76f61595f1a2cc8778ed7c279b6ca2e1fd19df3", size = 5587844, upload-time = "2025-11-06T22:07:45.033Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/17/b9/67240254166ae1eaa38dec32265e9153ac53645a6c6670ed36ad00722af8/ruff-0.14.4-py3-none-linux_armv6l.whl", hash = "sha256:e6604613ffbcf2297cd5dcba0e0ac9bd0c11dc026442dfbb614504e87c349518", size = 12606781, upload-time = "2025-11-06T22:07:01.841Z" },
+    { url = "https://files.pythonhosted.org/packages/46/c8/09b3ab245d8652eafe5256ab59718641429f68681ee713ff06c5c549f156/ruff-0.14.4-py3-none-macosx_10_12_x86_64.whl", hash = "sha256:d99c0b52b6f0598acede45ee78288e5e9b4409d1ce7f661f0fa36d4cbeadf9a4", size = 12946765, upload-time = "2025-11-06T22:07:05.858Z" },
+    { url = "https://files.pythonhosted.org/packages/14/bb/1564b000219144bf5eed2359edc94c3590dd49d510751dad26202c18a17d/ruff-0.14.4-py3-none-macosx_11_0_arm64.whl", hash = "sha256:9358d490ec030f1b51d048a7fd6ead418ed0826daf6149e95e30aa67c168af33", size = 11928120, upload-time = "2025-11-06T22:07:08.023Z" },
+    { url = "https://files.pythonhosted.org/packages/a3/92/d5f1770e9988cc0742fefaa351e840d9aef04ec24ae1be36f333f96d5704/ruff-0.14.4-py3-none-manylinux_2_17_aarch64.manylinux2014_aarch64.whl", hash = "sha256:81b40d27924f1f02dfa827b9c0712a13c0e4b108421665322218fc38caf615c2", size = 12370877, upload-time = "2025-11-06T22:07:10.015Z" },
+    { url = "https://files.pythonhosted.org/packages/e2/29/e9282efa55f1973d109faf839a63235575519c8ad278cc87a182a366810e/ruff-0.14.4-py3-none-manylinux_2_17_armv7l.manylinux2014_armv7l.whl", hash = "sha256:f5e649052a294fe00818650712083cddc6cc02744afaf37202c65df9ea52efa5", size = 12408538, upload-time = "2025-11-06T22:07:13.085Z" },
+    { url = "https://files.pythonhosted.org/packages/8e/01/930ed6ecfce130144b32d77d8d69f5c610e6d23e6857927150adf5d7379a/ruff-0.14.4-py3-none-manylinux_2_17_i686.manylinux2014_i686.whl", hash = "sha256:aa082a8f878deeba955531f975881828fd6afd90dfa757c2b0808aadb437136e", size = 13141942, upload-time = "2025-11-06T22:07:15.386Z" },
+    { url = "https://files.pythonhosted.org/packages/6a/46/a9c89b42b231a9f487233f17a89cbef9d5acd538d9488687a02ad288fa6b/ruff-0.14.4-py3-none-manylinux_2_17_ppc64.manylinux2014_ppc64.whl", hash = "sha256:1043c6811c2419e39011890f14d0a30470f19d47d197c4858b2787dfa698f6c8", size = 14544306, upload-time = "2025-11-06T22:07:17.631Z" },
+    { url = "https://files.pythonhosted.org/packages/78/96/9c6cf86491f2a6d52758b830b89b78c2ae61e8ca66b86bf5a20af73d20e6/ruff-0.14.4-py3-none-manylinux_2_17_ppc64le.manylinux2014_ppc64le.whl", hash = "sha256:a9f3a936ac27fb7c2a93e4f4b943a662775879ac579a433291a6f69428722649", size = 14210427, upload-time = "2025-11-06T22:07:19.832Z" },
+    { url = "https://files.pythonhosted.org/packages/71/f4/0666fe7769a54f63e66404e8ff698de1dcde733e12e2fd1c9c6efb689cb5/ruff-0.14.4-py3-none-manylinux_2_17_s390x.manylinux2014_s390x.whl", hash = "sha256:95643ffd209ce78bc113266b88fba3d39e0461f0cbc8b55fb92505030fb4a850", size = 13658488, upload-time = "2025-11-06T22:07:22.32Z" },
+    { url = "https://files.pythonhosted.org/packages/ee/79/6ad4dda2cfd55e41ac9ed6d73ef9ab9475b1eef69f3a85957210c74ba12c/ruff-0.14.4-py3-none-manylinux_2_17_x86_64.manylinux2014_x86_64.whl", hash = "sha256:456daa2fa1021bc86ca857f43fe29d5d8b3f0e55e9f90c58c317c1dcc2afc7b5", size = 13354908, upload-time = "2025-11-06T22:07:24.347Z" },
+    { url = "https://files.pythonhosted.org/packages/b5/60/f0b6990f740bb15c1588601d19d21bcc1bd5de4330a07222041678a8e04f/ruff-0.14.4-py3-none-manylinux_2_31_riscv64.whl", hash = "sha256:f911bba769e4a9f51af6e70037bb72b70b45a16db5ce73e1f72aefe6f6d62132", size = 13587803, upload-time = "2025-11-06T22:07:26.327Z" },
+    { url = "https://files.pythonhosted.org/packages/c9/da/eaaada586f80068728338e0ef7f29ab3e4a08a692f92eb901a4f06bbff24/ruff-0.14.4-py3-none-musllinux_1_2_aarch64.whl", hash = "sha256:76158a7369b3979fa878612c623a7e5430c18b2fd1c73b214945c2d06337db67", size = 12279654, upload-time = "2025-11-06T22:07:28.46Z" },
+    { url = "https://files.pythonhosted.org/packages/66/d4/b1d0e82cf9bf8aed10a6d45be47b3f402730aa2c438164424783ac88c0ed/ruff-0.14.4-py3-none-musllinux_1_2_armv7l.whl", hash = "sha256:f3b8f3b442d2b14c246e7aeca2e75915159e06a3540e2f4bed9f50d062d24469", size = 12357520, upload-time = "2025-11-06T22:07:31.468Z" },
+    { url = "https://files.pythonhosted.org/packages/04/f4/53e2b42cc82804617e5c7950b7079d79996c27e99c4652131c6a1100657f/ruff-0.14.4-py3-none-musllinux_1_2_i686.whl", hash = "sha256:c62da9a06779deecf4d17ed04939ae8b31b517643b26370c3be1d26f3ef7dbde", size = 12719431, upload-time = "2025-11-06T22:07:33.831Z" },
+    { url = "https://files.pythonhosted.org/packages/a2/94/80e3d74ed9a72d64e94a7b7706b1c1ebaa315ef2076fd33581f6a1cd2f95/ruff-0.14.4-py3-none-musllinux_1_2_x86_64.whl", hash = "sha256:5a443a83a1506c684e98acb8cb55abaf3ef725078be40237463dae4463366349", size = 13464394, upload-time = "2025-11-06T22:07:35.905Z" },
+    { url = "https://files.pythonhosted.org/packages/54/1a/a49f071f04c42345c793d22f6cf5e0920095e286119ee53a64a3a3004825/ruff-0.14.4-py3-none-win32.whl", hash = "sha256:643b69cb63cd996f1fc7229da726d07ac307eae442dd8974dbc7cf22c1e18fff", size = 12493429, upload-time = "2025-11-06T22:07:38.43Z" },
+    { url = "https://files.pythonhosted.org/packages/bc/22/e58c43e641145a2b670328fb98bc384e20679b5774258b1e540207580266/ruff-0.14.4-py3-none-win_amd64.whl", hash = "sha256:26673da283b96fe35fa0c939bf8411abec47111644aa9f7cfbd3c573fb125d2c", size = 13635380, upload-time = "2025-11-06T22:07:40.496Z" },
+    { url = "https://files.pythonhosted.org/packages/30/bd/4168a751ddbbf43e86544b4de8b5c3b7be8d7167a2a5cb977d274e04f0a1/ruff-0.14.4-py3-none-win_arm64.whl", hash = "sha256:dd09c292479596b0e6fec8cd95c65c3a6dc68e9ad17b8f2382130f87ff6a75bb", size = 12663065, upload-time = "2025-11-06T22:07:42.603Z" },
+]
+
 [[package]]
 name = "setuptools"
 version = "80.9.0"
@@ -1867,6 +2074,18 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/e9/44/75a9c9421471a6c4805dbf2356f7c181a29c1879239abab1ea2cc8f38b40/sniffio-1.3.1-py3-none-any.whl", hash = "sha256:2f6da418d1f1e0fddd844478f41680e794e6051915791a034ff65e5f100525a2", size = 10235, upload-time = "2024-02-25T23:20:01.196Z" },
 ]
 
+[[package]]
+name = "sse-starlette"
+version = "3.0.3"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "anyio" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/db/3c/fa6517610dc641262b77cc7bf994ecd17465812c1b0585fe33e11be758ab/sse_starlette-3.0.3.tar.gz", hash = "sha256:88cfb08747e16200ea990c8ca876b03910a23b547ab3bd764c0d8eb81019b971", size = 21943, upload-time = "2025-10-30T18:44:20.117Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/23/a0/984525d19ca5c8a6c33911a0c164b11490dd0f90ff7fd689f704f84e9a11/sse_starlette-3.0.3-py3-none-any.whl", hash = "sha256:af5bf5a6f3933df1d9c7f8539633dc8444ca6a97ab2e2a7cd3b6e431ac03a431", size = 11765, upload-time = "2025-10-30T18:44:18.834Z" },
+]
+
 [[package]]
 name = "starlette"
 version = "0.47.3"
@@ -1880,6 +2099,20 @@ wheels = [
     { url = "https://files.pythonhosted.org/packages/ce/fd/901cfa59aaa5b30a99e16876f11abe38b59a1a2c51ffb3d7142bb6089069/starlette-0.47.3-py3-none-any.whl", hash = "sha256:89c0778ca62a76b826101e7c709e70680a1699ca7da6b44d38eb0a7e61fe4b51", size = 72991, upload-time = "2025-08-24T13:36:40.887Z" },
 ]
 
+[[package]]
+name = "tavily-python"
+version = "0.7.12"
+source = { registry = "https://pypi.org/simple" }
+dependencies = [
+    { name = "httpx" },
+    { name = "requests" },
+    { name = "tiktoken" },
+]
+sdist = { url = "https://files.pythonhosted.org/packages/3e/42/ce2329635b844dda548110a5dfa0ab5631cdc1085e15c2d68b1850a2d112/tavily_python-0.7.12.tar.gz", hash = "sha256:661945bbc9284cdfbe70fb50de3951fd656bfd72e38e352481d333a36ae91f5a", size = 17282, upload-time = "2025-09-10T17:02:01.281Z" }
+wheels = [
+    { url = "https://files.pythonhosted.org/packages/9a/e2/dbc246d9fb24433f77b17d9ee4e750a1e2718432ebde2756589c9154cbad/tavily_python-0.7.12-py3-none-any.whl", hash = "sha256:00d09b9de3ca02ef9a994cf4e7ae43d4ec9d199f0566ba6e52cbfcbd07349bd1", size = 15473, upload-time = "2025-09-10T17:01:59.859Z" },
+]
+
 [[package]]
 name = "tenacity"
 version = "9.1.2"

docker-compose.yml

@@ -20,17 +20,22 @@ services:
     volumes:
       - ./backend/app:/app/app
       - ./backend/.secrets:/app/.secrets
+      - ./backend/data:/app/data
     env_file:
       - .env
     networks:
       - app-network
 
   db:
-    # docker run -p 6379:6379 -p 8001:8001 redis/redis-stack
     image: redis/redis-stack:latest
     ports:
       - 6379:6379
       - 8001:8001
+    environment:
+        REDIS_ARGS: --appendonly yes
+    volumes:
+      - ./redis_data:/data
+    restart: unless-stopped
     networks:
       - app-network
 

frontend/package-lock.json

@@ -8,6 +8,7 @@
       "name": "frontend",
       "version": "0.0.0",
       "dependencies": {
+        "@ai-sdk/react": "^2.0.89",
         "@radix-ui/react-checkbox": "^1.3.3",
         "@radix-ui/react-collapsible": "^1.1.12",
         "@radix-ui/react-dialog": "^1.1.15",
@@ -110,6 +111,30 @@
         "zod": "^3.25.76 || ^4.1.8"
       }
     },
+    "node_modules/@ai-sdk/react": {
+      "version": "2.0.89",
+      "resolved": "https://registry.npmjs.org/@ai-sdk/react/-/react-2.0.89.tgz",
+      "integrity": "sha512-r2uCqx042JOjNrSlDrjh7ufSIfU2BM6Lo4qe47KHkYuJjPfssxhLpJUCFLB01iV7Foyn/xpbq06Zr6WI4qUDgw==",
+      "license": "Apache-2.0",
+      "dependencies": {
+        "@ai-sdk/provider-utils": "3.0.16",
+        "ai": "5.0.89",
+        "swr": "^2.2.5",
+        "throttleit": "2.1.0"
+      },
+      "engines": {
+        "node": ">=18"
+      },
+      "peerDependencies": {
+        "react": "^18 || ^19 || ^19.0.0-rc",
+        "zod": "^3.25.76 || ^4.1.8"
+      },
+      "peerDependenciesMeta": {
+        "zod": {
+          "optional": true
+        }
+      }
+    },
     "node_modules/@alloc/quick-lru": {
       "version": "5.2.0",
       "resolved": "https://registry.npmjs.org/@alloc/quick-lru/-/quick-lru-5.2.0.tgz",
@@ -8818,6 +8843,19 @@
         "node": ">=8"
       }
     },
+    "node_modules/swr": {
+      "version": "2.3.6",
+      "resolved": "https://registry.npmjs.org/swr/-/swr-2.3.6.tgz",
+      "integrity": "sha512-wfHRmHWk/isGNMwlLGlZX5Gzz/uTgo0o2IRuTMcf4CPuPFJZlq0rDaKUx+ozB5nBOReNV1kiOyzMfj+MBMikLw==",
+      "license": "MIT",
+      "dependencies": {
+        "dequal": "^2.0.3",
+        "use-sync-external-store": "^1.4.0"
+      },
+      "peerDependencies": {
+        "react": "^16.11.0 || ^17.0.0 || ^18.0.0 || ^19.0.0"
+      }
+    },
     "node_modules/tailwind-merge": {
       "version": "3.3.1",
       "resolved": "https://registry.npmjs.org/tailwind-merge/-/tailwind-merge-3.3.1.tgz",
@@ -8857,6 +8895,18 @@
         "url": "https://opencollective.com/webpack"
       }
     },
+    "node_modules/throttleit": {
+      "version": "2.1.0",
+      "resolved": "https://registry.npmjs.org/throttleit/-/throttleit-2.1.0.tgz",
+      "integrity": "sha512-nt6AMGKW1p/70DF/hGBdJB57B8Tspmbp5gfJ8ilhLnt7kkr2ye7hzD6NVG8GGErk2HWF34igrL2CXmNIkzKqKw==",
+      "license": "MIT",
+      "engines": {
+        "node": ">=18"
+      },
+      "funding": {
+        "url": "https://github.com/sponsors/sindresorhus"
+      }
+    },
     "node_modules/tiny-invariant": {
       "version": "1.3.3",
       "resolved": "https://registry.npmjs.org/tiny-invariant/-/tiny-invariant-1.3.3.tgz",

frontend/package.json

@@ -10,6 +10,7 @@
     "preview": "vite preview"
   },
   "dependencies": {
+    "@ai-sdk/react": "^2.0.89",
     "@radix-ui/react-checkbox": "^1.3.3",
     "@radix-ui/react-collapsible": "^1.1.12",
     "@radix-ui/react-dialog": "^1.1.15",

frontend/src/components/AnalystReport.tsx

@@ -0,0 +1,230 @@
+import {
+  TrendingUp,
+  TrendingDown,
+  Activity,
+  Info,
+  Target,
+  Lightbulb,
+  ArrowRight,
+  AlertTriangle,
+} from "lucide-react";
+
+type TrendDirection = "Improving" | "Declining" | "Stable" | "Volatile";
+
+interface TrendMetricPoint {
+  year: number;
+  value: number;
+  growth?: number | null;
+}
+
+interface TrendMetric {
+  name: string;
+  unit: string;
+  description: string;
+  points: TrendMetricPoint[];
+  cagr?: number | null;
+  direction: TrendDirection;
+  notes?: string | null;
+}
+
+interface TrendInsight {
+  category: string;
+  direction: TrendDirection;
+  summary: string;
+  confidence: number;
+}
+
+interface AnalystReportData {
+  organisation_name: string;
+  organisation_ein: string;
+  years_analyzed: number[];
+  key_metrics: TrendMetric[];
+  insights: TrendInsight[];
+  recommendations: string[];
+  outlook: string;
+}
+
+interface AnalystReportProps {
+  data: AnalystReportData;
+}
+
+const directionBadgeClasses: Record<TrendDirection, string> = {
+  Improving: "bg-green-100 text-green-700 border-green-200",
+  Declining: "bg-red-100 text-red-700 border-red-200",
+  Stable: "bg-blue-100 text-blue-700 border-blue-200",
+  Volatile: "bg-yellow-100 text-yellow-700 border-yellow-200",
+};
+
+const directionIcon = (direction: TrendDirection) => {
+  switch (direction) {
+    case "Improving":
+      return <TrendingUp className="w-4 h-4" />;
+    case "Declining":
+      return <TrendingDown className="w-4 h-4" />;
+    case "Volatile":
+      return <Activity className="w-4 h-4" />;
+    default:
+      return <Info className="w-4 h-4" />;
+  }
+};
+
+const formatNumber = (value: number, unit: string) => {
+  if (unit === "USD") {
+    return `$${value.toLocaleString("en-US", {
+      maximumFractionDigits: 0,
+    })}`;
+  }
+  if (unit === "Ratio") {
+    return `${(value * 100).toFixed(1)}%`;
+  }
+  return value.toLocaleString("en-US", { maximumFractionDigits: 2 });
+};
+
+const formatPercent = (value?: number | null) => {
+  if (value === undefined || value === null || Number.isNaN(value)) {
+    return "—";
+  }
+  return `${(value * 100).toFixed(1)}%`;
+};
+
+export function AnalystReport({ data }: AnalystReportProps) {
+  const {
+    organisation_name,
+    organisation_ein,
+    years_analyzed,
+    key_metrics,
+    insights,
+    recommendations,
+    outlook,
+  } = data;
+
+  return (
+    <div className="w-full bg-white border border-gray-200 rounded-lg shadow-sm overflow-hidden">
+      <div className="bg-gradient-to-r from-emerald-50 to-sky-50 px-4 py-3 border-b border-emerald-100">
+        <div className="flex items-center justify-between">
+          <div>
+            <p className="text-xs text-gray-500 uppercase tracking-wide">
+              Multi-year performance analysis
+            </p>
+            <h3 className="text-lg font-semibold text-gray-900">
+              {organisation_name}
+            </h3>
+            <p className="text-xs text-gray-600">
+              EIN {organisation_ein} • {years_analyzed.join(" – ")}
+            </p>
+          </div>
+          <div className="text-right text-sm text-gray-600">
+            <span className="font-medium text-gray-900">Outlook</span>
+            <p className="text-xs text-gray-600 max-w-xs">{outlook}</p>
+          </div>
+        </div>
+      </div>
+
+      {/* Key Metrics */}
+      {key_metrics.length > 0 && (
+        <div className="p-4 space-y-3">
+          <h4 className="text-sm font-semibold text-gray-800 flex items-center gap-2">
+            <Activity className="w-4 h-4 text-emerald-600" />
+            Core Trend Metrics
+          </h4>
+          <div className="grid gap-3 md:grid-cols-2">
+            {key_metrics.slice(0, 4).map((metric, index) => {
+              const latest = metric.points[metric.points.length - 1];
+              const prior =
+                metric.points.length > 1
+                  ? metric.points[metric.points.length - 2]
+                  : null;
+              const yoy = latest && prior ? formatPercent(latest.growth) : "—";
+              return (
+                <div
+                  key={`${metric.name}-${index}`}
+                  className="border rounded-lg p-3 bg-gray-50"
+                >
+                  <div className="flex items-center justify-between mb-1">
+                    <div className="font-medium text-sm text-gray-900">
+                      {metric.name}
+                    </div>
+                    <span
+                      className={`inline-flex items-center gap-1 text-xs px-2 py-0.5 rounded-full border ${directionBadgeClasses[metric.direction]}`}
+                    >
+                      {directionIcon(metric.direction)}
+                      {metric.direction}
+                    </span>
+                  </div>
+                  <p className="text-2xl font-semibold text-gray-900">
+                    {latest ? formatNumber(latest.value, metric.unit) : "—"}
+                  </p>
+                  <div className="flex justify-between text-xs text-gray-600 mt-1">
+                    <span>YoY: {yoy}</span>
+                    <span>CAGR: {formatPercent(metric.cagr)}</span>
+                  </div>
+                  {metric.notes && (
+                    <p className="text-xs text-gray-500 mt-2">{metric.notes}</p>
+                  )}
+                </div>
+              );
+            })}
+          </div>
+        </div>
+      )}
+
+      {/* Insights */}
+      {insights.length > 0 && (
+        <div className="px-4 pb-4">
+          <h4 className="text-sm font-semibold text-gray-800 mb-2 flex items-center gap-2">
+            <Lightbulb className="w-4 h-4 text-amber-500" />
+            Key Insights
+          </h4>
+          <div className="space-y-2">
+            {insights.map((insight, index) => (
+              <div
+                key={`${insight.category}-${index}`}
+                className="border rounded-lg p-3 bg-white"
+              >
+                <div className="flex justify-between items-center mb-1">
+                  <div className="flex items-center gap-2 text-sm font-medium text-gray-900">
+                    {directionIcon(insight.direction)}
+                    {insight.category}
+                  </div>
+                  <span className="text-xs text-gray-500">
+                    {Math.round(insight.confidence * 100)}% confidence
+                  </span>
+                </div>
+                <p className="text-sm text-gray-700">{insight.summary}</p>
+              </div>
+            ))}
+          </div>
+        </div>
+      )}
+
+      {/* Recommendations */}
+      {recommendations.length > 0 && (
+        <div className="px-4 pb-4">
+          <h4 className="text-sm font-semibold text-gray-800 mb-2 flex items-center gap-2">
+            <Target className="w-4 h-4 text-indigo-500" />
+            Recommended Actions
+          </h4>
+          <ul className="space-y-1 text-sm text-gray-700">
+            {recommendations.map((rec, index) => (
+              <li
+                key={`rec-${index}`}
+                className="flex items-start gap-2 bg-gray-50 border border-gray-100 rounded-lg p-2"
+              >
+                <ArrowRight className="w-4 h-4 text-gray-500 mt-0.5" />
+                <span>{rec}</span>
+              </li>
+            ))}
+          </ul>
+        </div>
+      )}
+
+      {/* Empty states fallback */}
+      {recommendations.length === 0 && insights.length === 0 && (
+        <div className="px-4 py-6 text-sm text-gray-600 flex items-center gap-2">
+          <AlertTriangle className="w-4 h-4 text-gray-400" />
+          No trend insights available yet. Try requesting an annual comparison.
+        </div>
+      )}
+    </div>
+  );
+}

frontend/src/components/AuditReport.tsx

@@ -0,0 +1,326 @@
+import React, { useState } from "react";
+import {
+  AlertTriangle,
+  CheckCircle,
+  XCircle,
+  FileText,
+  AlertCircle,
+  ChevronDown,
+  ChevronRight,
+  Shield,
+  Info,
+} from "lucide-react";
+
+type Severity = "Pass" | "Warning" | "Error";
+
+interface AuditFinding {
+  check_id: string;
+  category: string;
+  severity: Severity;
+  message: string;
+  mitigation?: string;
+  confidence: number;
+}
+
+interface AuditSectionSummary {
+  section: string;
+  severity: Severity;
+  summary: string;
+  confidence: number;
+}
+
+interface AuditReportData {
+  organisation_ein: string;
+  organisation_name: string;
+  year?: number;
+  overall_severity: Severity;
+  findings: AuditFinding[];
+  sections: AuditSectionSummary[];
+  overall_summary?: string;
+  notes?: string;
+}
+
+interface AuditReportProps {
+  data: AuditReportData;
+}
+
+const getSeverityIcon = (severity: Severity, size = "w-4 h-4") => {
+  switch (severity) {
+    case "Pass":
+      return <CheckCircle className={`${size} text-green-600`} />;
+    case "Warning":
+      return <AlertTriangle className={`${size} text-yellow-600`} />;
+    case "Error":
+      return <XCircle className={`${size} text-red-600`} />;
+    default:
+      return <AlertCircle className={`${size} text-gray-600`} />;
+  }
+};
+
+const getSeverityBadge = (severity: Severity) => {
+  const colors = {
+    Pass: "bg-green-100 text-green-800 border-green-200",
+    Warning: "bg-yellow-100 text-yellow-800 border-yellow-200",
+    Error: "bg-red-100 text-red-800 border-red-200",
+  };
+
+  return (
+    <span
+      className={`inline-flex items-center gap-1 px-2 py-1 rounded-full text-xs font-medium border ${colors[severity]}`}
+    >
+      {getSeverityIcon(severity, "w-3 h-3")}
+      {severity}
+    </span>
+  );
+};
+
+export const AuditReport: React.FC<AuditReportProps> = ({ data }) => {
+  const [expandedSections, setExpandedSections] = useState<Set<string>>(
+    new Set(),
+  );
+  const [showAllFindings, setShowAllFindings] = useState(false);
+
+  const {
+    organisation_name,
+    organisation_ein,
+    year,
+    overall_severity,
+    findings,
+    sections,
+    overall_summary,
+    notes,
+  } = data;
+
+  const severityStats = {
+    Pass: findings.filter((f) => f.severity === "Pass").length,
+    Warning: findings.filter((f) => f.severity === "Warning").length,
+    Error: findings.filter((f) => f.severity === "Error").length,
+  };
+
+  const toggleSection = (section: string) => {
+    const newExpanded = new Set(expandedSections);
+    if (newExpanded.has(section)) {
+      newExpanded.delete(section);
+    } else {
+      newExpanded.add(section);
+    }
+    setExpandedSections(newExpanded);
+  };
+
+  const criticalFindings = findings.filter((f) => f.severity === "Error");
+
+  return (
+    <div className="w-full bg-white border border-gray-200 rounded-lg shadow-sm overflow-hidden">
+      {/* Compact Header */}
+      <div className="bg-linear-to-r from-blue-50 to-indigo-50 p-4">
+        <div className="flex items-center justify-between">
+          <div className="flex items-center gap-3">
+            <Shield className="w-6 h-6 text-blue-600" />
+            <div>
+              <h3 className="font-semibold text-gray-900">
+                {organisation_name}
+              </h3>
+              <div className="flex items-center gap-3 text-xs text-gray-600">
+                <span>EIN: {organisation_ein}</span>
+                {year && <span>{year}</span>}
+              </div>
+            </div>
+          </div>
+          {getSeverityBadge(overall_severity)}
+        </div>
+      </div>
+
+      {/* Quick Stats */}
+      <div className="bg-gray-50 px-4 py-3 border-b">
+        <div className="flex items-center justify-between">
+          <span className="text-sm font-medium text-gray-700">
+            Audit Results
+          </span>
+          <div className="flex items-center gap-4 text-sm">
+            {severityStats.Pass > 0 && (
+              <div className="flex items-center gap-1">
+                <CheckCircle className="w-3 h-3 text-green-600" />
+                <span className="font-medium text-green-700">
+                  {severityStats.Pass}
+                </span>
+              </div>
+            )}
+            {severityStats.Warning > 0 && (
+              <div className="flex items-center gap-1">
+                <AlertTriangle className="w-3 h-3 text-yellow-600" />
+                <span className="font-medium text-yellow-700">
+                  {severityStats.Warning}
+                </span>
+              </div>
+            )}
+            {severityStats.Error > 0 && (
+              <div className="flex items-center gap-1">
+                <XCircle className="w-3 h-3 text-red-600" />
+                <span className="font-medium text-red-700">
+                  {severityStats.Error}
+                </span>
+              </div>
+            )}
+          </div>
+        </div>
+      </div>
+
+      <div className="p-4 space-y-4">
+        {/* Summary */}
+        {overall_summary && (
+          <div className="bg-blue-50 border border-blue-200 rounded-lg p-3">
+            <div className="flex items-start gap-2">
+              <Info className="w-4 h-4 text-blue-600 mt-0.5 shrink-0" />
+              <div>
+                <h4 className="font-medium text-blue-900 text-sm mb-1">
+                  Summary
+                </h4>
+                <p className="text-sm text-blue-800">{overall_summary}</p>
+              </div>
+            </div>
+          </div>
+        )}
+
+        {/* Critical Issues (if any) */}
+        {criticalFindings.length > 0 && (
+          <div className="bg-red-50 border border-red-200 rounded-lg p-3">
+            <h4 className="font-medium text-red-900 text-sm mb-2 flex items-center gap-2">
+              <XCircle className="w-4 h-4" />
+              Critical Issues ({criticalFindings.length})
+            </h4>
+            <div className="space-y-2">
+              {criticalFindings.slice(0, 2).map((finding, index) => (
+                <div key={index} className="text-sm text-red-800">
+                  <span className="font-medium">{finding.category}:</span>{" "}
+                  {finding.message}
+                </div>
+              ))}
+              {criticalFindings.length > 2 && (
+                <button
+                  onClick={() => setShowAllFindings(!showAllFindings)}
+                  className="text-xs text-red-700 hover:text-red-800 font-medium"
+                >
+                  {showAllFindings
+                    ? "Show less"
+                    : `+${criticalFindings.length - 2} more issues`}
+                </button>
+              )}
+            </div>
+          </div>
+        )}
+
+        {/* Sections Overview */}
+        {sections.length > 0 && (
+          <div>
+            <button
+              onClick={() => toggleSection("sections")}
+              className="flex items-center gap-2 w-full text-left p-2 hover:bg-gray-50 rounded-lg"
+            >
+              {expandedSections.has("sections") ? (
+                <ChevronDown className="w-4 h-4" />
+              ) : (
+                <ChevronRight className="w-4 h-4" />
+              )}
+              <span className="font-medium text-sm">
+                Section Analysis ({sections.length})
+              </span>
+            </button>
+
+            {expandedSections.has("sections") && (
+              <div className="mt-2 grid gap-2 sm:grid-cols-2">
+                {sections.map((section, index) => (
+                  <div key={index} className="border rounded-lg p-3 bg-gray-50">
+                    <div className="flex items-center justify-between mb-1">
+                      <span className="font-medium text-sm">
+                        {section.section}
+                      </span>
+                      <div className="flex items-center gap-1">
+                        {getSeverityIcon(section.severity)}
+                        <span className="text-xs text-gray-600">
+                          {Math.round(section.confidence * 100)}%
+                        </span>
+                      </div>
+                    </div>
+                    <p className="text-xs text-gray-700">{section.summary}</p>
+                  </div>
+                ))}
+              </div>
+            )}
+          </div>
+        )}
+
+        {/* All Findings */}
+        <div>
+          <button
+            onClick={() => toggleSection("findings")}
+            className="flex items-center gap-2 w-full text-left p-2 hover:bg-gray-50 rounded-lg"
+          >
+            {expandedSections.has("findings") ? (
+              <ChevronDown className="w-4 h-4" />
+            ) : (
+              <ChevronRight className="w-4 h-4" />
+            )}
+            <span className="font-medium text-sm">
+              All Findings ({findings.length})
+            </span>
+          </button>
+
+          {expandedSections.has("findings") && (
+            <div className="mt-2 space-y-2">
+              {findings.map((finding, index) => (
+                <div key={index} className="border rounded-lg p-3 bg-gray-50">
+                  <div className="flex items-start justify-between mb-2">
+                    <div className="flex items-center gap-2">
+                      {getSeverityIcon(finding.severity)}
+                      <div>
+                        <span className="font-medium text-sm">
+                          {finding.category}
+                        </span>
+                        <span className="text-xs text-gray-500 ml-1">
+                          #{finding.check_id}
+                        </span>
+                      </div>
+                    </div>
+                    <span className="text-xs text-gray-600">
+                      {Math.round(finding.confidence * 100)}% confidence
+                    </span>
+                  </div>
+
+                  <p className="text-sm text-gray-700 mb-2">
+                    {finding.message}
+                  </p>
+
+                  {finding.mitigation && (
+                    <div className="bg-white rounded p-2 border">
+                      <span className="text-xs font-medium text-gray-600">
+                        Recommended:
+                      </span>
+                      <p className="text-xs text-gray-700 mt-1">
+                        {finding.mitigation}
+                      </p>
+                    </div>
+                  )}
+                </div>
+              ))}
+            </div>
+          )}
+        </div>
+
+        {/* Notes */}
+        {notes && (
+          <div className="border-t pt-3">
+            <div className="flex items-start gap-2">
+              <FileText className="w-4 h-4 text-gray-400 mt-0.5 shrink-0" />
+              <div>
+                <h4 className="font-medium text-gray-700 text-sm mb-1">
+                  Notes
+                </h4>
+                <p className="text-sm text-gray-600 italic">{notes}</p>
+              </div>
+            </div>
+          </div>
+        )}
+      </div>
+    </div>
+  );
+};

frontend/src/components/ChatTab.tsx

@@ -1,92 +1,412 @@
-import { MessageCircle, Send, Bot, User } from "lucide-react";
+import { Message, MessageContent } from "@/components/ai-elements/message";
-import { Button } from "@/components/ui/button";
+import {
-import { Input } from "@/components/ui/input";
+  PromptInput,
+  PromptInputActionAddAttachments,
+  PromptInputActionMenu,
+  PromptInputActionMenuContent,
+  PromptInputActionMenuTrigger,
+  PromptInputAttachment,
+  PromptInputAttachments,
+  PromptInputBody,
+  PromptInputHeader,
+  type PromptInputMessage,
+  PromptInputSubmit,
+  PromptInputTextarea,
+  PromptInputFooter,
+  PromptInputTools,
+} from "@/components/ai-elements/prompt-input";
+import { Action, Actions } from "@/components/ai-elements/actions";
+import { Fragment, useState, useEffect, useMemo } from "react";
+import { useChat } from "@ai-sdk/react";
+import { Response } from "@/components/ai-elements/response";
+import {
+  CopyIcon,
+  RefreshCcwIcon,
+  MessageCircle,
+  Bot,
+  AlertCircle,
+  PaperclipIcon,
+  User,
+} from "lucide-react";
+import { AuditReport } from "./AuditReport";
+import { AnalystReport } from "./AnalystReport";
+import { WebSearchResults } from "./WebSearchResults";
+import { Loader } from "@/components/ai-elements/loader";
+import { DefaultChatTransport } from "ai";
 
 interface ChatTabProps {
   selectedTema: string | null;
 }
 
 export function ChatTab({ selectedTema }: ChatTabProps) {
+  const [input, setInput] = useState("");
+  const [error, setError] = useState<string | null>(null);
+
+  const {
+    messages,
+    sendMessage,
+    status,
+    regenerate,
+    error: chatError,
+  } = useChat({
+    transport: new DefaultChatTransport({
+      api: "/api/v1/agent/chat",
+      headers: {
+        tema: selectedTema || "",
+      },
+    }),
+    onError: (error) => {
+      setError(`Chat error: ${error.message}`);
+    },
+  });
+
+  // Clear error when starting new conversation
+  useEffect(() => {
+    if (status === "streaming") {
+      setError(null);
+    }
+  }, [status]);
+
+  const handleSubmit = (message: PromptInputMessage) => {
+    const hasText = Boolean(message.text?.trim());
+    const hasAttachments = Boolean(message.files?.length);
+
+    if (!(hasText || hasAttachments)) {
+      return;
+    }
+
+    setError(null);
+    sendMessage(
+      {
+        text: message.text || "Enviado con archivos adjuntos",
+        files: message.files,
+      },
+      {
+        body: {
+          dataroom: selectedTema,
+          context: `User is asking about the dataroom: ${selectedTema}`,
+        },
+      },
+    );
+    setInput("");
+  };
+
+  const hasActiveToolRequest = useMemo(() => {
+    return messages.some((message) =>
+      message.parts.some(
+        (part: any) =>
+          typeof part?.type === "string" &&
+          part.type.startsWith("tool-") &&
+          part.state === "input-available",
+      ),
+    );
+  }, [messages]);
+
+  const shouldShowGlobalLoader =
+    (status === "streaming" || status === "loading") && !hasActiveToolRequest;
+
   if (!selectedTema) {
     return (
       <div className="flex flex-col items-center justify-center h-64">
         <MessageCircle className="w-12 h-12 text-gray-400 mb-4" />
-        <p className="text-gray-500">
+        <p className="text-gray-500">Select a dataroom to start chatting</p>
-          Selecciona un dataroom para iniciar el chat
-        </p>
       </div>
     );
   }
 
   return (
-    <div className="flex flex-col h-full">
+    <div className="flex flex-col h-[638px] max-h-[638px]">
-      {/* Chat Header */}
+      {/* Chat Content */}
-      <div className="border-b border-gray-200 px-6 py-4">
+      <div className="flex-1 min-h-0 overflow-y-auto">
-        <div className="flex items-center gap-3">
+        <div className="max-w-4xl mx-auto w-full space-y-6 p-6">
-          <div className="p-2 bg-blue-100 rounded-lg">
-            <MessageCircle className="w-5 h-5 text-blue-600" />
-          </div>
-          <div>
-            <h3 className="text-lg font-semibold text-gray-900">
-              Chat con {selectedTema}
-            </h3>
-            <p className="text-sm text-gray-600">
-              Haz preguntas sobre los documentos de este dataroom
-            </p>
-          </div>
-        </div>
-      </div>
-
-      {/* Chat Messages Area */}
-      <div className="flex-1 overflow-y-auto p-6">
-        <div className="max-w-4xl mx-auto space-y-4">
           {/* Welcome Message */}
-          <div className="flex items-start gap-3">
+          {messages.length === 0 && (
+            <div className="flex items-start gap-3 mb-6">
               <div className="p-2 bg-blue-100 rounded-full">
                 <Bot className="w-4 h-4 text-blue-600" />
               </div>
               <div className="flex-1 bg-gray-50 rounded-lg p-4">
                 <p className="text-sm text-gray-800">
-                ¡Hola! Soy tu asistente de IA para el dataroom <strong>{selectedTema}</strong>.
+                  Hi! I’m your AI assistant for dataroom{" "}
-                Puedes hacerme preguntas sobre los documentos almacenados aquí.
+                  <strong>{selectedTema}</strong>. Ask me anything about the
+                  stored documents.
                 </p>
               </div>
             </div>
+          )}
 
-          {/* Placeholder for future messages */}
+          {/* Error Message */}
-          <div className="text-center py-8">
+          {error && (
-            <MessageCircle className="w-16 h-16 text-gray-300 mx-auto mb-4" />
+            <div className="flex items-start gap-3 mb-4">
-            <h4 className="text-lg font-medium text-gray-900 mb-2">
+              <div className="p-2 bg-red-100 rounded-full">
-              Chat Inteligente
+                <AlertCircle className="w-4 h-4 text-red-600" />
-            </h4>
+              </div>
-            <p className="text-gray-500 max-w-md mx-auto">
+              <div className="flex-1 bg-red-50 rounded-lg p-4 border border-red-200">
-              El chat estará disponible próximamente. Podrás hacer preguntas sobre los
+                <p className="text-sm text-red-800">{error}</p>
-              documentos y obtener respuestas basadas en el contenido del dataroom.
-            </p>
-          </div>
               </div>
             </div>
+          )}
 
-      {/* Chat Input Area */}
+          {/* Chat Messages */}
-      <div className="border-t border-gray-200 p-6">
+          {messages.map((message) => (
-        <div className="max-w-4xl mx-auto">
+            <div key={message.id}>
-          <div className="flex items-center gap-3">
+              {message.parts.map((part, i) => {
+                switch (part.type) {
+                  case "text":
+                    return (
+                      <Fragment key={`${message.id}-${i}`}>
+                        {message.role === "user" ? (
+                          <div className="flex items-start gap-3 justify-end">
                             <div className="flex-1">
-              <Input
+                              <Message
-                placeholder={`Pregunta algo sobre ${selectedTema}...`}
+                                from={message.role}
-                disabled
+                                className="max-w-none"
-                className="w-full"
+                              >
-              />
+                                <MessageContent>
+                                  <Response>{part.text}</Response>
+                                </MessageContent>
+                              </Message>
                             </div>
-            <Button disabled className="gap-2">
+                            <div className="p-2 rounded-full flex-shrink-0 mt-1 bg-gray-100">
-              <Send className="w-4 h-4" />
+                              <User className="w-4 h-4 text-gray-600" />
-              Enviar
-            </Button>
                             </div>
-          <p className="text-xs text-gray-500 mt-2">
+                          </div>
-            Esta funcionalidad estará disponible próximamente
+                        ) : (
+                          <div className="flex items-start gap-3">
+                            <div className="p-2 rounded-full flex-shrink-0 mt-1 bg-blue-100">
+                              <Bot className="w-4 h-4 text-blue-600" />
+                            </div>
+                            <div className="flex-1">
+                              <Message
+                                from={message.role}
+                                className="max-w-none"
+                              >
+                                <MessageContent>
+                                  <Response>{part.text}</Response>
+                                </MessageContent>
+                              </Message>
+                            </div>
+                          </div>
+                        )}
+                        {message.role === "assistant" &&
+                          i === message.parts.length - 1 && (
+                            <div className="ml-12">
+                              <Actions className="mt-2">
+                                <Action
+                                  onClick={() => regenerate()}
+                                  label="Regenerar"
+                                  disabled={status === "streaming"}
+                                >
+                                  <RefreshCcwIcon className="size-3" />
+                                </Action>
+                                <Action
+                                  onClick={() =>
+                                    navigator.clipboard.writeText(part.text)
+                                  }
+                                  label="Copiar"
+                                >
+                                  <CopyIcon className="size-3" />
+                                </Action>
+                              </Actions>
+                            </div>
+                          )}
+                      </Fragment>
+                    );
+                  case "tool-build_audit_report":
+                    switch (part.state) {
+                      case "input-available":
+                        return (
+                          <div
+                            key={`${message.id}-${i}`}
+                            className="mb-4 flex items-center gap-2 p-4 bg-blue-50 rounded-lg border border-blue-200"
+                          >
+                            <div className="animate-spin rounded-full h-4 w-4 border-b-2 border-blue-600"></div>
+                            <span className="text-sm text-blue-700">
+                              Generating audit report…
+                            </span>
+                          </div>
+                        );
+                      case "output-available":
+                        return (
+                          <div
+                            key={`${message.id}-${i}`}
+                            className="mt-4 mb-4 w-full"
+                          >
+                            <div className="max-w-full overflow-hidden">
+                              <AuditReport data={part.output} />
+                            </div>
+                          </div>
+                        );
+                      case "output-error":
+                        return (
+                          <div
+                            key={`${message.id}-${i}`}
+                            className="mb-4 p-4 bg-red-50 border border-red-200 rounded-lg"
+                          >
+                            <div className="flex items-center gap-2">
+                              <AlertCircle className="w-4 h-4 text-red-600" />
+                              <span className="text-sm font-medium text-red-800">
+                                Failed to generate audit report
+                              </span>
+                            </div>
+                            <p className="text-sm text-red-600 mt-1">
+                              {part.errorText}
                             </p>
                           </div>
+                        );
+                      default:
+                        return null;
+                    }
+                  case "tool-build_analysis_report":
+                    switch (part.state) {
+                      case "input-available":
+                        return (
+                          <div
+                            key={`${message.id}-${i}`}
+                            className="mb-4 flex items-center gap-2 p-4 bg-purple-50 rounded-lg border border-purple-200"
+                          >
+                            <div className="animate-spin rounded-full h-4 w-4 border-b-2 border-purple-600"></div>
+                            <span className="text-sm text-purple-700">
+                              Generating performance analysis…
+                            </span>
+                          </div>
+                        );
+                      case "output-available":
+                        return (
+                          <div
+                            key={`${message.id}-${i}`}
+                            className="mt-4 mb-4 w-full"
+                          >
+                            <div className="max-w-full overflow-hidden">
+                              <AnalystReport data={part.output} />
+                            </div>
+                          </div>
+                        );
+                      case "output-error":
+                        return (
+                          <div
+                            key={`${message.id}-${i}`}
+                            className="mb-4 p-4 bg-red-50 border border-red-200 rounded-lg"
+                          >
+                            <div className="flex items-center gap-2">
+                              <AlertCircle className="w-4 h-4 text-red-600" />
+                              <span className="text-sm font-medium text-red-800">
+                                Failed to generate performance analysis
+                              </span>
+                            </div>
+                            <p className="text-sm text-red-600 mt-1">
+                              {part.errorText}
+                            </p>
+                          </div>
+                        );
+                      default:
+                        return null;
+                    }
+                  case "tool-search_web_information":
+                    switch (part.state) {
+                      case "input-available":
+                        return (
+                          <div
+                            key={`${message.id}-${i}`}
+                            className="mb-4 flex items-center gap-2 p-4 bg-green-50 rounded-lg border border-green-200"
+                          >
+                            <div className="animate-spin rounded-full h-4 w-4 border-b-2 border-green-600"></div>
+                            <span className="text-sm text-green-700">
+                              Searching the web…
+                            </span>
+                          </div>
+                        );
+                      case "output-available":
+                        return (
+                          <div
+                            key={`${message.id}-${i}`}
+                            className="mt-4 mb-4 w-full"
+                          >
+                            <div className="max-w-full overflow-hidden">
+                              <WebSearchResults data={part.output} />
+                            </div>
+                          </div>
+                        );
+                      case "output-error":
+                        return (
+                          <div
+                            key={`${message.id}-${i}`}
+                            className="mb-4 p-4 bg-red-50 border border-red-200 rounded-lg"
+                          >
+                            <div className="flex items-center gap-2">
+                              <AlertCircle className="w-4 h-4 text-red-600" />
+                              <span className="text-sm font-medium text-red-800">
+                                Failed to search the web
+                              </span>
+                            </div>
+                            <p className="text-sm text-red-600 mt-1">
+                              {part.errorText}
+                            </p>
+                          </div>
+                        );
+                      default:
+                        return null;
+                    }
+                  default:
+                    return null;
+                }
+              })}
+            </div>
+          ))}
+          {shouldShowGlobalLoader && <Loader />}
+        </div>
+      </div>
+
+      {/* Chat Input */}
+      <div className="border-t border-gray-200 p-3 bg-gray-50/50 flex-shrink-0">
+        <PromptInput
+          onSubmit={handleSubmit}
+          className="max-w-4xl mx-auto border-2 border-gray-200 rounded-xl focus-within:border-slate-500 transition-colors duration-200 bg-white"
+          globalDrop
+          multiple
+        >
+          <PromptInputHeader className="p-2 pb-0">
+            <PromptInputAttachments>
+              {(attachment) => <PromptInputAttachment data={attachment} />}
+            </PromptInputAttachments>
+          </PromptInputHeader>
+          <PromptInputBody>
+            <PromptInputTextarea
+              onChange={(e) => setInput(e.target.value)}
+              value={input}
+              placeholder={`Ask something about ${selectedTema}...`}
+              disabled={status === "streaming" || status === "loading"}
+              className="min-h-[60px] resize-none border-0 focus:ring-0 transition-all duration-200 text-base px-4 py-3 bg-white rounded-xl"
+            />
+          </PromptInputBody>
+          <PromptInputFooter className="mt-3 flex justify-between items-center">
+            <PromptInputTools>
+              <PromptInputActionMenu>
+                <PromptInputActionMenuTrigger>
+                  <PaperclipIcon className="size-4" />
+                </PromptInputActionMenuTrigger>
+                <PromptInputActionMenuContent>
+                  <PromptInputActionAddAttachments />
+                </PromptInputActionMenuContent>
+              </PromptInputActionMenu>
+            </PromptInputTools>
+            <PromptInputSubmit
+              disabled={
+                (!input.trim() && !status) ||
+                status === "streaming" ||
+                status === "loading"
+              }
+              status={status}
+              className={`rounded-full px-6 py-2 font-medium transition-all duration-200 flex items-center gap-2 ${
+                (!input.trim() && !status) ||
+                status === "streaming" ||
+                status === "loading"
+                  ? "bg-gray-300 cursor-not-allowed text-gray-500"
+                  : "bg-blue-600 hover:bg-blue-700 text-white"
+              }`}
+            />
+          </PromptInputFooter>
+        </PromptInput>
       </div>
     </div>
   );

frontend/src/components/DashboardTab.tsx

@@ -1,32 +1,134 @@
-import { FileText, Users, Database, Activity } from "lucide-react";
+import { useState, useEffect } from "react";
+import {
+  FileText,
+  Database,
+  Activity,
+  TrendingUp,
+  AlertCircle,
+  CheckCircle,
+  Loader2,
+} from "lucide-react";
+import { api } from "@/services/api";
 
 interface DashboardTabProps {
   selectedTema: string | null;
 }
 
+interface DataroomInfo {
+  name: string;
+  collection: string;
+  storage: string;
+  file_count: number;
+  total_size_bytes: number;
+  total_size_mb: number;
+  collection_exists: boolean;
+  vector_count: number | null;
+  collection_info: {
+    vectors_count: number;
+    indexed_vectors_count: number;
+    points_count: number;
+    segments_count: number;
+    status: string;
+  } | null;
+  file_types: Record<string, number>;
+  recent_files: Array<{
+    name: string;
+    size_mb: number;
+    last_modified: string;
+  }>;
+}
+
 export function DashboardTab({ selectedTema }: DashboardTabProps) {
+  const [dataroomInfo, setDataroomInfo] = useState<DataroomInfo | null>(null);
+  const [loading, setLoading] = useState(false);
+  const [error, setError] = useState<string | null>(null);
+
+  useEffect(() => {
+    if (selectedTema) {
+      fetchDataroomInfo();
+    }
+  }, [selectedTema]);
+
+  const fetchDataroomInfo = async () => {
+    if (!selectedTema) return;
+
+    setLoading(true);
+    setError(null);
+
+    try {
+      const info = await api.getDataroomInfo(selectedTema);
+      setDataroomInfo(info);
+    } catch (err) {
+      const errorMessage = err instanceof Error ? err.message : "Unknown error";
+      setError(`Unable to load dataroom info: ${errorMessage}`);
+      console.error("Error fetching dataroom info:", err);
+    } finally {
+      setLoading(false);
+    }
+  };
+
+  const formatFileTypes = (fileTypes: Record<string, number>) => {
+    const entries = Object.entries(fileTypes);
+    if (entries.length === 0) return "No files";
+
+    return entries
+      .sort(([, a], [, b]) => b - a) // Sort by count descending
+      .slice(0, 3) // Take top 3
+      .map(([ext, count]) => `${ext.toUpperCase()}: ${count}`)
+      .join(", ");
+  };
+
+  const formatBytes = (bytes: number) => {
+    if (bytes === 0) return "0 MB";
+    const mb = bytes / (1024 * 1024);
+    if (mb < 1) return `${(bytes / 1024).toFixed(1)} KB`;
+    return `${mb.toFixed(1)} MB`;
+  };
+
   if (!selectedTema) {
     return (
       <div className="flex flex-col items-center justify-center h-64">
         <Activity className="w-12 h-12 text-gray-400 mb-4" />
-        <p className="text-gray-500">
+        <p className="text-gray-500">Select a dataroom to view its metrics</p>
-          Selecciona un dataroom para ver las métricas
+      </div>
-        </p>
+    );
+  }
+
+  if (loading) {
+    return (
+      <div className="flex flex-col items-center justify-center h-64">
+        <Loader2 className="w-8 h-8 text-blue-600 animate-spin mb-4" />
+        <p className="text-gray-600">Loading metrics…</p>
+      </div>
+    );
+  }
+
+  if (error) {
+    return (
+      <div className="p-6">
+        <div className="bg-red-50 border border-red-200 rounded-lg p-4 flex items-center gap-3">
+          <AlertCircle className="w-5 h-5 text-red-600 flex-shrink-0" />
+          <div>
+            <p className="text-sm font-medium text-red-800">Error</p>
+            <p className="text-sm text-red-600">{error}</p>
+          </div>
+        </div>
+      </div>
+    );
+  }
+
+  if (!dataroomInfo) {
+    return (
+      <div className="flex flex-col items-center justify-center h-64">
+        <AlertCircle className="w-12 h-12 text-gray-400 mb-4" />
+        <p className="text-gray-500">Unable to load dataroom information</p>
       </div>
     );
   }
 
   return (
     <div className="p-6">
-      <div className="mb-6">
+      <h4 className="text-md font-semibold text-gray-900 mb-4">Metrics</h4>
-        <h3 className="text-lg font-semibold text-gray-900 mb-2">
-          Métricas del Dataroom: {selectedTema}
-        </h3>
-        <p className="text-sm text-gray-600">
-          Vista general del estado y actividad del dataroom
-        </p>
-      </div>
-
       <div className="grid grid-cols-1 md:grid-cols-2 lg:grid-cols-4 gap-6">
         {/* Files Count Card */}
         <div className="bg-white border border-gray-200 rounded-lg p-4">
@@ -35,8 +137,13 @@ export function DashboardTab({ selectedTema }: DashboardTabProps) {
               <FileText className="w-5 h-5 text-blue-600" />
             </div>
             <div>
-              <p className="text-sm font-medium text-gray-600">Archivos</p>
+              <p className="text-sm font-medium text-gray-600">Files</p>
-              <p className="text-2xl font-bold text-gray-900">--</p>
+              <p className="text-2xl font-bold text-gray-900">
+                {dataroomInfo.file_count}
+              </p>
+              <p className="text-xs text-gray-500 mt-1">
+                {formatFileTypes(dataroomInfo.file_types)}
+              </p>
             </div>
           </div>
         </div>
@@ -48,8 +155,13 @@ export function DashboardTab({ selectedTema }: DashboardTabProps) {
               <Database className="w-5 h-5 text-green-600" />
             </div>
             <div>
-              <p className="text-sm font-medium text-gray-600">Almacenamiento</p>
+              <p className="text-sm font-medium text-gray-600">Storage</p>
-              <p className="text-2xl font-bold text-gray-900">--</p>
+              <p className="text-2xl font-bold text-gray-900">
+                {dataroomInfo.total_size_mb.toFixed(1)} MB
+              </p>
+              <p className="text-xs text-gray-500 mt-1">
+                {formatBytes(dataroomInfo.total_size_bytes)}
+              </p>
             </div>
           </div>
         </div>
@@ -61,38 +173,98 @@ export function DashboardTab({ selectedTema }: DashboardTabProps) {
               <Activity className="w-5 h-5 text-purple-600" />
             </div>
             <div>
-              <p className="text-sm font-medium text-gray-600">Vectores</p>
+              <p className="text-sm font-medium text-gray-600">Vectors</p>
-              <p className="text-2xl font-bold text-gray-900">--</p>
+              <p className="text-2xl font-bold text-gray-900">
-            </div>
+                {dataroomInfo.vector_count ?? 0}
-          </div>
+              </p>
-        </div>
+              <p className="text-xs text-gray-500 mt-1">
-
+                {dataroomInfo.collection_exists
-        {/* Activity Card */}
+                  ? "Indexed vectors"
-        <div className="bg-white border border-gray-200 rounded-lg p-4">
+                  : "No vectors"}
-          <div className="flex items-center gap-3">
-            <div className="p-2 bg-orange-100 rounded-lg">
-              <Users className="w-5 h-5 text-orange-600" />
-            </div>
-            <div>
-              <p className="text-sm font-medium text-gray-600">Actividad</p>
-              <p className="text-2xl font-bold text-gray-900">--</p>
-            </div>
-          </div>
-        </div>
-      </div>
-
-      {/* Coming Soon Message */}
-      <div className="mt-8 bg-gray-50 border border-gray-200 rounded-lg p-6">
-        <div className="text-center">
-          <Activity className="w-8 h-8 text-gray-400 mx-auto mb-3" />
-          <h4 className="text-sm font-medium text-gray-900 mb-2">
-            Panel de Métricas
-          </h4>
-          <p className="text-sm text-gray-500">
-            Este panel se llenará con métricas detalladas y gráficos interactivos próximamente.
               </p>
             </div>
           </div>
         </div>
+
+        {/* Collection Status Card */}
+        <div className="bg-white border border-gray-200 rounded-lg p-4">
+          <div className="flex items-center gap-3">
+            <div className="p-2 bg-orange-100 rounded-lg">
+              <TrendingUp className="w-5 h-5 text-orange-600" />
+            </div>
+            <div>
+              <p className="text-sm font-medium text-gray-600">Status</p>
+              <div className="flex items-center gap-2">
+                <p className="text-2xl font-bold text-gray-900">
+                  {dataroomInfo.collection_exists ? "Active" : "Inactive"}
+                </p>
+                {dataroomInfo.collection_exists ? (
+                  <CheckCircle className="w-6 h-6 text-green-600" />
+                ) : (
+                  <AlertCircle className="w-6 h-6 text-yellow-600" />
+                )}
+              </div>
+              {dataroomInfo.collection_info ? (
+                <p className="text-xs text-gray-500 mt-1">
+                  {dataroomInfo.collection_info.indexed_vectors_count}/
+                  {dataroomInfo.collection_info.vectors_count} indexed vectors
+                </p>
+              ) : (
+                <p className="text-xs text-gray-500 mt-1">
+                  {dataroomInfo.collection_exists
+                    ? "Collection has no data"
+                    : "No collection"}
+                </p>
+              )}
+            </div>
+          </div>
+        </div>
+      </div>
+
+      {/* Recent Files Section */}
+      {dataroomInfo.recent_files.length > 0 && (
+        <div className="mt-8">
+          <h4 className="text-md font-semibold text-gray-900 mb-4">
+            Recent Files
+          </h4>
+          <div className="bg-white border border-gray-200 rounded-lg overflow-hidden">
+            <div className="divide-y divide-gray-200">
+              {dataroomInfo.recent_files.map((file, index) => (
+                <div
+                  key={index}
+                  className="p-4 flex items-center justify-between hover:bg-gray-50"
+                >
+                  <div className="flex items-center gap-3">
+                    <FileText className="w-4 h-4 text-gray-400" />
+                    <div>
+                      <p className="text-sm font-medium text-gray-900">
+                        {file.name}
+                      </p>
+                      <p className="text-xs text-gray-500">
+                        {new Date(file.last_modified).toLocaleDateString(
+                          "en-US",
+                          {
+                            year: "numeric",
+                            month: "short",
+                            day: "numeric",
+                            hour: "2-digit",
+                            minute: "2-digit",
+                          },
+                        )}
+                      </p>
+                    </div>
+                  </div>
+                  <div className="text-right">
+                    <p className="text-sm text-gray-600">
+                      {file.size_mb.toFixed(2)} MB
+                    </p>
+                  </div>
+                </div>
+              ))}
+            </div>
+          </div>
+        </div>
+      )}
+    </div>
   );
 }

frontend/src/components/DataroomView.tsx

@@ -1,158 +1,111 @@
-import { useEffect, useState } from "react";
+import { useState } from "react";
 import { useFileStore } from "@/stores/fileStore";
-import { api } from "@/services/api";
 import { Tabs, TabsContent, TabsList, TabsTrigger } from "@/components/ui/tabs";
+import {
+  Dialog,
+  DialogContent,
+  DialogHeader,
+  DialogTitle,
+} from "@/components/ui/dialog";
+import { Button } from "@/components/ui/button";
+import { Expand, Minimize2 } from "lucide-react";
 import { FilesTab } from "./FilesTab";
 import { DashboardTab } from "./DashboardTab";
 import { ChatTab } from "./ChatTab";
-import {
-  CheckCircle2,
-  AlertCircle,
-  Loader2,
-} from "lucide-react";
 
 interface DataroomViewProps {
   onProcessingChange?: (isProcessing: boolean) => void;
 }
 
 export function DataroomView({ onProcessingChange }: DataroomViewProps = {}) {
-  const { selectedTema, files } = useFileStore();
+  const { selectedTema } = useFileStore();
 
-  // Collection status states
-  const [isCheckingCollection, setIsCheckingCollection] = useState(false);
-  const [collectionExists, setCollectionExists] = useState<boolean | null>(
-    null,
-  );
-  const [collectionError, setCollectionError] = useState<string | null>(null);
   const [processing, setProcessing] = useState(false);
-
+  const [fullscreenTab, setFullscreenTab] = useState<string | null>(null);
-  // Check collection status when tema changes
+  const [currentTab, setCurrentTab] = useState("overview");
-  useEffect(() => {
-    checkCollectionStatus();
-  }, [selectedTema]);
-
-  // Load files when tema changes
-  useEffect(() => {
-    loadFiles();
-  }, [selectedTema]);
-
-  const checkCollectionStatus = async () => {
-    if (!selectedTema) {
-      setCollectionExists(null);
-      return;
-    }
-
-    setIsCheckingCollection(true);
-    setCollectionError(null);
-
-    try {
-      const result = await api.checkCollectionExists(selectedTema);
-      setCollectionExists(result.exists);
-    } catch (err) {
-      console.error("Error checking collection:", err);
-      setCollectionError(
-        err instanceof Error ? err.message : "Error al verificar colección",
-      );
-      setCollectionExists(null);
-    } finally {
-      setIsCheckingCollection(false);
-    }
-  };
-
-  const handleCreateCollection = async () => {
-    if (!selectedTema) return;
-
-    setIsCheckingCollection(true);
-    setCollectionError(null);
-
-    try {
-      const result = await api.createCollection(selectedTema);
-      if (result.success) {
-        setCollectionExists(true);
-        console.log(`Collection "${selectedTema}" created successfully`);
-      }
-    } catch (err) {
-      console.error("Error creating collection:", err);
-      setCollectionError(
-        err instanceof Error ? err.message : "Error al crear colección",
-      );
-    } finally {
-      setIsCheckingCollection(false);
-    }
-  };
-
-  const loadFiles = async () => {
-    // This will be handled by FilesTab component
-  };
 
   const handleProcessingChange = (isProcessing: boolean) => {
     setProcessing(isProcessing);
     onProcessingChange?.(isProcessing);
   };
 
-  const totalFiles = files.length;
+  const openFullscreen = (tabValue: string) => {
+    setFullscreenTab(tabValue);
+  };
+
+  const closeFullscreen = () => {
+    setFullscreenTab(null);
+  };
+
+  const renderTabContent = (tabValue: string, isFullscreen = false) => {
+    const className = isFullscreen ? "h-[calc(100vh-8rem)] flex flex-col" : "";
+
+    switch (tabValue) {
+      case "overview":
+        return (
+          <div className={className}>
+            <DashboardTab selectedTema={selectedTema} />
+          </div>
+        );
+      case "files":
+        return (
+          <div className={className}>
+            <FilesTab
+              selectedTema={selectedTema}
+              processing={processing}
+              onProcessingChange={handleProcessingChange}
+            />
+          </div>
+        );
+      case "chat":
+        return (
+          <div className={className}>
+            <ChatTab selectedTema={selectedTema} />
+          </div>
+        );
+      default:
+        return null;
+    }
+  };
+
+  const getTabTitle = (tabValue: string) => {
+    switch (tabValue) {
+      case "overview":
+        return "Overview";
+      case "files":
+        return "Files";
+      case "chat":
+        return "Chat";
+      default:
+        return "";
+    }
+  };
 
   return (
     <div className="flex flex-col h-full bg-white">
       <div className="border-b border-gray-200 px-6 py-4">
         <div className="flex flex-wrap items-center justify-between gap-4">
           <div>
-            <div className="flex items-center gap-3 mb-2">
+            <h2 className="text-2xl font-semibold text-gray-900 mb-2">
-              <h2 className="text-2xl font-semibold text-gray-900">
+              {selectedTema ? `Dataroom: ${selectedTema}` : "Select a dataroom"}
-                {selectedTema
-                  ? `Dataroom: ${selectedTema}`
-                  : "Selecciona un dataroom"}
             </h2>
-              {/* Collection Status Indicator */}
-              {selectedTema && (
-                <div className="flex items-center gap-2">
-                  {isCheckingCollection ? (
-                    <>
-                      <Loader2 className="w-4 h-4 animate-spin text-gray-500" />
-                      <span className="text-xs text-gray-500">
-                        Verificando...
-                      </span>
-                    </>
-                  ) : collectionExists === true ? (
-                    <>
-                      <CheckCircle2 className="w-4 h-4 text-green-600" />
-                      <span className="text-xs text-green-600">
-                        Colección disponible
-                      </span>
-                    </>
-                  ) : collectionExists === false ? (
-                    <>
-                      <AlertCircle className="w-4 h-4 text-yellow-600" />
-                      <button
-                        onClick={handleCreateCollection}
-                        className="text-xs text-yellow-600 hover:text-yellow-700 underline"
-                      >
-                        Crear colección
-                      </button>
-                    </>
-                  ) : collectionError ? (
-                    <>
-                      <AlertCircle className="w-4 h-4 text-red-600" />
-                      <span className="text-xs text-red-600">
-                        Error de conexión
-                      </span>
-                    </>
-                  ) : null}
-                </div>
-              )}
-            </div>
             <p className="text-sm text-gray-600">
               {selectedTema
-                ? `${totalFiles} archivo${totalFiles !== 1 ? "s" : ""}`
+                ? "Manage files, review metrics, and chat with AI about the content."
-                : "Selecciona un dataroom de la barra lateral para ver sus archivos"}
+                : "Pick a dataroom from the sidebar to get started."}
             </p>
           </div>
         </div>
       </div>
 
-      <Tabs defaultValue="files" className="flex flex-col flex-1">
+      <Tabs
+        value={currentTab}
+        onValueChange={setCurrentTab}
+        className="flex flex-col flex-1"
+      >
         <div className="border-b border-gray-200 px-6 py-2">
-          <TabsList className="flex h-10 w-full items-center gap-2 bg-transparent p-0 justify-start">
+          <TabsList className="flex h-10 w-full items-center gap-2 bg-transparent p-0 justify-between">
+            <div className="flex items-center gap-2">
               <TabsTrigger
                 value="overview"
                 className="rounded-md px-4 py-2 text-sm font-medium text-gray-600 transition data-[state=active]:bg-gray-900 data-[state=active]:text-white data-[state=active]:shadow"
@@ -171,25 +124,60 @@ export function DataroomView({ onProcessingChange }: DataroomViewProps = {}) {
               >
                 Chat
               </TabsTrigger>
+            </div>
+
+            <Button
+              variant="outline"
+              size="sm"
+              onClick={() => openFullscreen(currentTab)}
+              className="ml-auto"
+            >
+              <Expand className="h-4 w-4" />
+              <span className="sr-only">Open fullscreen</span>
+            </Button>
           </TabsList>
         </div>
 
         <TabsContent value="overview" className="mt-0 flex-1">
-          <DashboardTab selectedTema={selectedTema} />
+          {renderTabContent("overview")}
         </TabsContent>
 
         <TabsContent value="files" className="mt-0 flex flex-1 flex-col">
-          <FilesTab
+          {renderTabContent("files")}
-            selectedTema={selectedTema}
-            processing={processing}
-            onProcessingChange={handleProcessingChange}
-          />
         </TabsContent>
 
         <TabsContent value="chat" className="mt-0 flex-1">
-          <ChatTab selectedTema={selectedTema} />
+          {renderTabContent("chat")}
         </TabsContent>
       </Tabs>
+
+      <Dialog
+        open={fullscreenTab !== null}
+        onOpenChange={(open: boolean) => !open && closeFullscreen()}
+      >
+        <DialogContent className="max-w-[100vw] max-h-[100vh] w-[100vw] h-[100vh] m-0 rounded-none [&>button]:hidden">
+          <DialogHeader className="flex flex-row items-center justify-between space-y-0 pb-4">
+            <DialogTitle className="text-xl font-semibold">
+              {selectedTema
+                ? `${getTabTitle(fullscreenTab || "")} - ${selectedTema}`
+                : getTabTitle(fullscreenTab || "")}
+            </DialogTitle>
+            <Button
+              variant="outline"
+              size="sm"
+              onClick={closeFullscreen}
+              className="h-8 w-8 p-0"
+            >
+              <Minimize2 className="h-4 w-4" />
+              <span className="sr-only">Exit fullscreen</span>
+            </Button>
+          </DialogHeader>
+
+          <div className="flex-1 overflow-hidden">
+            {fullscreenTab && renderTabContent(fullscreenTab, true)}
+          </div>
+        </DialogContent>
+      </Dialog>
     </div>
   );
 }

frontend/src/components/DeleteConfirmDialog.tsx

@@ -1,4 +1,4 @@
-import { Button } from '@/components/ui/button'
+import { Button } from "@/components/ui/button";
 import {
   Dialog,
   DialogContent,
@@ -6,17 +6,17 @@ import {
   DialogFooter,
   DialogHeader,
   DialogTitle,
-} from '@/components/ui/dialog'
+} from "@/components/ui/dialog";
-import { Trash2, AlertTriangle } from 'lucide-react'
+import { Trash2, AlertTriangle } from "lucide-react";
 
 interface DeleteConfirmDialogProps {
-  open: boolean
+  open: boolean;
-  onOpenChange: (open: boolean) => void
+  onOpenChange: (open: boolean) => void;
-  onConfirm: () => void
+  onConfirm: () => void;
-  title: string
+  title: string;
-  description: string
+  description: string;
-  fileList?: string[]
+  fileList?: string[];
-  loading?: boolean
+  loading?: boolean;
 }
 
 export function DeleteConfirmDialog({
@@ -26,7 +26,7 @@ export function DeleteConfirmDialog({
   title,
   description,
   fileList,
-  loading = false
+  loading = false,
 }: DeleteConfirmDialogProps) {
   return (
     <Dialog open={open} onOpenChange={onOpenChange}>
@@ -41,7 +41,7 @@ export function DeleteConfirmDialog({
 
         {fileList && fileList.length > 0 && (
           <div className="max-h-40 overflow-y-auto bg-gray-50 rounded p-3">
-            <p className="text-sm font-medium mb-2">Archivos a eliminar:</p>
+            <p className="text-sm	font-medium mb-2">Files to delete:</p>
             <ul className="text-sm space-y-1">
               {fileList.map((filename, index) => (
                 <li key={index} className="flex items-center gap-2">
@@ -59,17 +59,13 @@ export function DeleteConfirmDialog({
             onClick={() => onOpenChange(false)}
             disabled={loading}
           >
-            Cancelar
+            Cancel
           </Button>
-          <Button
+          <Button variant="destructive" onClick={onConfirm} disabled={loading}>
-            variant="destructive"
+            {loading ? "Deleting…" : "Delete"}
-            onClick={onConfirm}
-            disabled={loading}
-          >
-            {loading ? 'Eliminando...' : 'Eliminar'}
           </Button>
         </DialogFooter>
       </DialogContent>
     </Dialog>
-  )
+  );
 }

frontend/src/components/FilesTab.tsx

@@ -1,4 +1,4 @@
-import { useState } from "react";
+import { useState, useEffect } from "react";
 import { useFileStore } from "@/stores/fileStore";
 import { api } from "@/services/api";
 import { Button } from "@/components/ui/button";
@@ -60,7 +60,7 @@ export function FilesTab({
   const [deleting, setDeleting] = useState(false);
   const [downloading, setDownloading] = useState(false);
 
-  // Estados para el modal de preview de PDF
+  // PDF preview modal state
   const [previewModalOpen, setPreviewModalOpen] = useState(false);
   const [previewFileUrl, setPreviewFileUrl] = useState<string | null>(null);
   const [previewFileName, setPreviewFileName] = useState("");
@@ -69,17 +69,22 @@ export function FilesTab({
   );
   const [loadingPreview, setLoadingPreview] = useState(false);
 
-  // Estados para el modal de chunks
+  // Chunk viewer modal state
   const [chunkViewerOpen, setChunkViewerOpen] = useState(false);
   const [chunkFileName, setChunkFileName] = useState("");
   const [chunkFileTema, setChunkFileTema] = useState("");
 
-  // Estados para chunking
+  // LandingAI chunking state
   const [chunkingConfigOpen, setChunkingConfigOpen] = useState(false);
   const [chunkingFileName, setChunkingFileName] = useState("");
   const [chunkingFileTema, setChunkingFileTema] = useState("");
   const [chunkingCollectionName, setChunkingCollectionName] = useState("");
 
+  // Load files when component mounts or selectedTema changes
+  useEffect(() => {
+    loadFiles();
+  }, [selectedTema]);
+
   const loadFiles = async () => {
     // Don't load files if no dataroom is selected
     if (!selectedTema) {
@@ -118,10 +123,10 @@ export function FilesTab({
       setDeleting(true);
 
       if (fileToDelete) {
-        // Eliminar archivo individual
+        // Delete single file
         await api.deleteFile(fileToDelete, selectedTema || undefined);
       } else {
-        // Eliminar archivos seleccionados
+        // Delete selected files
         const filesToDelete = Array.from(selectedFiles);
         await api.deleteFiles(filesToDelete, selectedTema || undefined);
         clearSelection();
@@ -154,9 +159,7 @@ export function FilesTab({
     try {
       setDownloading(true);
       const filesToDownload = Array.from(selectedFiles);
-      const zipName = selectedTema
+      const zipName = selectedTema ? `${selectedTema}_files` : "selected_files";
-        ? `${selectedTema}_archivos`
-        : "archivos_seleccionados";
       await api.downloadMultipleFiles(
         filesToDownload,
         selectedTema || undefined,
@@ -213,10 +216,10 @@ export function FilesTab({
         tema: chunkingFileTema,
         collection_name: chunkingCollectionName,
         mode: config.mode,
-        schema_id: config.schemaId,
+        schema_id: config.schema_id,
-        include_chunk_types: config.includeChunkTypes,
+        include_chunk_types: config.include_chunk_types,
-        max_tokens_per_chunk: config.maxTokensPerChunk,
+        max_tokens_per_chunk: config.max_tokens_per_chunk,
-        merge_small_chunks: config.mergeSmallChunks,
+        merge_small_chunks: config.merge_small_chunks,
       };
 
       await api.processWithLandingAI(processConfig);
@@ -229,7 +232,7 @@ export function FilesTab({
     }
   };
 
-  // Filtrar archivos por término de búsqueda
+  // Filter files by search term
   const filteredFiles = files.filter((file) =>
     file.name.toLowerCase().includes(searchTerm.toLowerCase()),
   );
@@ -245,7 +248,7 @@ export function FilesTab({
   };
 
   const formatDate = (dateString: string): string => {
-    return new Date(dateString).toLocaleDateString("es-ES", {
+    return new Date(dateString).toLocaleDateString("en-US", {
       year: "numeric",
       month: "short",
       day: "numeric",
@@ -257,15 +260,15 @@ export function FilesTab({
   const getDeleteDialogProps = () => {
     if (fileToDelete) {
       return {
-        title: "Eliminar archivo",
+        title: "Delete file",
-        message: `¿Estás seguro de que deseas eliminar el archivo "${fileToDelete}"?`,
+        message: `Are you sure you want to delete "${fileToDelete}"?`,
         fileList: [fileToDelete],
       };
     } else {
       const filesToDelete = Array.from(selectedFiles);
       return {
-        title: "Eliminar archivos seleccionados",
+        title: "Delete selected files",
-        message: `¿Estás seguro de que deseas eliminar ${filesToDelete.length} archivo${filesToDelete.length > 1 ? "s" : ""}?`,
+        message: `Are you sure you want to delete ${filesToDelete.length} file${filesToDelete.length > 1 ? "s" : ""}?`,
         fileList: filesToDelete,
       };
     }
@@ -275,9 +278,7 @@ export function FilesTab({
     return (
       <div className="flex flex-col items-center justify-center h-64">
         <FileText className="w-12 h-12 text-gray-400 mb-4" />
-        <p className="text-gray-500">
+        <p className="text-gray-500">Select a dataroom to view its files</p>
-          Selecciona un dataroom para ver sus archivos
-        </p>
       </div>
     );
   }
@@ -289,7 +290,7 @@ export function FilesTab({
           <div className="flex items-center gap-3">
             <div className="animate-spin rounded-full h-4 w-4 border-b-2 border-blue-600"></div>
             <span className="text-sm text-blue-800">
-              Procesando archivos con LandingAI...
+              Processing files with LandingAI…
             </span>
           </div>
         </div>
@@ -300,7 +301,7 @@ export function FilesTab({
           <div className="relative flex-1 max-w-md">
             <Search className="absolute left-3 top-1/2 transform -translate-y-1/2 text-gray-400 w-4 h-4" />
             <Input
-              placeholder="Buscar archivos..."
+              placeholder="Search files..."
               value={searchTerm}
               onChange={(e) => setSearchTerm(e.target.value)}
               className="pl-10"
@@ -319,7 +320,7 @@ export function FilesTab({
                   className="gap-2"
                 >
                   <Download className="w-4 h-4" />
-                  Descargar ({selectedFiles.size})
+                  Download ({selectedFiles.size})
                 </Button>
                 <Button
                   variant="outline"
@@ -329,7 +330,7 @@ export function FilesTab({
                   className="gap-2 text-red-600 hover:text-red-700"
                 >
                   <Trash2 className="w-4 h-4" />
-                  Eliminar ({selectedFiles.size})
+                  Delete ({selectedFiles.size})
                 </Button>
               </>
             )}
@@ -340,7 +341,7 @@ export function FilesTab({
               className="gap-2"
             >
               <Upload className="w-4 h-4" />
-              Subir archivo
+              Upload files
             </Button>
           </div>
         </div>
@@ -350,17 +351,17 @@ export function FilesTab({
         <div className="p-6">
           {loading ? (
             <div className="flex items-center justify-center h-64">
-              <p className="text-gray-500">Cargando archivos...</p>
+              <p className="text-gray-500">Loading files…</p>
             </div>
           ) : filteredFiles.length === 0 ? (
             <div className="flex flex-col items-center justify-center h-64">
               <FileText className="w-12 h-12 text-gray-400 mb-4" />
               <p className="text-gray-500">
                 {!selectedTema
-                  ? "Selecciona un dataroom para ver sus archivos"
+                  ? "Select a dataroom to view its files"
                   : searchTerm
-                    ? "No se encontraron archivos"
+                    ? "No files match your search"
-                    : "No hay archivos en este dataroom"}
+                    : "This dataroom has no files yet"}
               </p>
             </div>
           ) : (
@@ -382,10 +383,10 @@ export function FilesTab({
                       }}
                     />
                   </TableHead>
-                  <TableHead>Archivo</TableHead>
+                  <TableHead>File</TableHead>
-                  <TableHead>Tamaño</TableHead>
+                  <TableHead>Size</TableHead>
-                  <TableHead>Modificado</TableHead>
+                  <TableHead>Modified</TableHead>
-                  <TableHead className="text-right">Acciones</TableHead>
+                  <TableHead className="text-right">Actions</TableHead>
                 </TableRow>
               </TableHeader>
               <TableBody>
@@ -413,7 +414,7 @@ export function FilesTab({
                           onClick={() => handlePreviewFile(file.name)}
                           disabled={loadingPreview}
                           className="h-8 w-8 p-0"
-                          title="Vista previa"
+                          title="Preview"
                         >
                           <Eye className="w-4 h-4" />
                         </Button>
@@ -422,7 +423,7 @@ export function FilesTab({
                           size="sm"
                           onClick={() => handleViewChunks(file.name)}
                           className="h-8 w-8 p-0"
-                          title="Ver chunks"
+                          title="View chunks"
                         >
                           <MessageSquare className="w-4 h-4" />
                         </Button>
@@ -431,7 +432,7 @@ export function FilesTab({
                           size="sm"
                           onClick={() => handleStartChunking(file.name)}
                           className="h-8 w-8 p-0"
-                          title="Procesar con LandingAI"
+                          title="Process with LandingAI"
                         >
                           <Scissors className="w-4 h-4" />
                         </Button>
@@ -441,7 +442,7 @@ export function FilesTab({
                           onClick={() => handleDownloadFile(file.name)}
                           disabled={downloading}
                           className="h-8 w-8 p-0"
-                          title="Descargar"
+                          title="Download"
                         >
                           <Download className="w-4 h-4" />
                         </Button>
@@ -451,7 +452,7 @@ export function FilesTab({
                           onClick={() => handleDeleteFile(file.name)}
                           disabled={deleting}
                           className="h-8 w-8 p-0 text-red-600 hover:text-red-700 hover:bg-red-50"
-                          title="Eliminar"
+                          title="Delete"
                         >
                           <Trash2 className="w-4 h-4" />
                         </Button>
@@ -498,7 +499,7 @@ export function FilesTab({
         tema={chunkFileTema}
       />
 
-      {/* Modal de configuración de chunking con LandingAI */}
+      {/* LandingAI chunking config modal */}
       <ChunkingConfigModalLandingAI
         isOpen={chunkingConfigOpen}
         onClose={() => setChunkingConfigOpen(false)}

frontend/src/components/PDFPreviewModal.tsx

@@ -1,25 +1,20 @@
-import { useState, useEffect } from 'react'
+import { useState, useEffect } from "react";
 import {
   Dialog,
   DialogContent,
   DialogHeader,
   DialogTitle,
-  DialogDescription
+  DialogDescription,
-} from '@/components/ui/dialog'
+} from "@/components/ui/dialog";
-import { Button } from '@/components/ui/button'
+import { Button } from "@/components/ui/button";
-import {
+import { Download, Loader2, FileText, ExternalLink } from "lucide-react";
-  Download,
-  Loader2,
-  FileText,
-  ExternalLink
-} from 'lucide-react'
 
 interface PDFPreviewModalProps {
-  open: boolean
+  open: boolean;
-  onOpenChange: (open: boolean) => void
+  onOpenChange: (open: boolean) => void;
-  fileUrl: string | null
+  fileUrl: string | null;
-  fileName: string
+  fileName: string;
-  onDownload?: () => void
+  onDownload?: () => void;
 }
 
 export function PDFPreviewModal({
@@ -27,45 +22,40 @@ export function PDFPreviewModal({
   onOpenChange,
   fileUrl,
   fileName,
-  onDownload
+  onDownload,
 }: PDFPreviewModalProps) {
-  // Estado para manejar el loading del iframe
+  // Track iframe loading state
-  const [loading, setLoading] = useState(true)
+  const [loading, setLoading] = useState(true);
 
-  // Efecto para manejar el timeout del loading
+  // Hide loading if iframe never fires onLoad
   useEffect(() => {
     if (open && fileUrl) {
-      setLoading(true)
+      setLoading(true);
 
-      // Timeout para ocultar loading automáticamente después de 3 segundos
-      // Algunos iframes no disparan onLoad correctamente
       const timeout = setTimeout(() => {
-        setLoading(false)
+        setLoading(false);
-      }, 3000)
+      }, 3000);
 
-      return () => clearTimeout(timeout)
+      return () => clearTimeout(timeout);
     }
-  }, [open, fileUrl])
+  }, [open, fileUrl]);
 
-  // Manejar cuando el iframe termina de cargar
   const handleIframeLoad = () => {
-    setLoading(false)
+    setLoading(false);
-  }
+  };
 
-  // Abrir PDF en nueva pestaña
   const openInNewTab = () => {
     if (fileUrl) {
-      window.open(fileUrl, '_blank')
+      window.open(fileUrl, "_blank");
-    }
     }
+  };
 
-  // Reiniciar loading cuando cambia el archivo
   const handleOpenChange = (open: boolean) => {
     if (open) {
-      setLoading(true)
+      setLoading(true);
-    }
-    onOpenChange(open)
     }
+    onOpenChange(open);
+  };
 
   return (
     <Dialog open={open} onOpenChange={handleOpenChange}>
@@ -75,81 +65,68 @@ export function PDFPreviewModal({
             <FileText className="w-5 h-5" />
             {fileName}
           </DialogTitle>
-          <DialogDescription>
+          <DialogDescription>PDF preview</DialogDescription>
-            Vista previa del documento PDF
-          </DialogDescription>
         </DialogHeader>
 
-        {/* Barra de controles */}
+        {/* Controls */}
         <div className="flex items-center justify-between gap-4 px-6 py-3 border-b bg-gray-50">
           <div className="flex items-center gap-2">
             <Button
               variant="outline"
               size="sm"
               onClick={openInNewTab}
-              title="Abrir en nueva pestaña"
+              title="Open in new tab"
             >
               <ExternalLink className="w-4 h-4 mr-2" />
-              Abrir en pestaña nueva
+              Open in new tab
             </Button>
           </div>
 
-          {/* Botón de descarga */}
+          {/* Download button */}
           {onDownload && (
             <Button
               variant="outline"
               size="sm"
               onClick={onDownload}
-              title="Descargar archivo"
+              title="Download file"
             >
               <Download className="w-4 h-4 mr-2" />
-              Descargar
+              Download
             </Button>
           )}
         </div>
 
-        {/* Área de visualización del PDF con iframe */}
+        {/* PDF iframe */}
-        <div className="flex-1 relative bg-gray-100">
+        <div className="flex-1 relative bg-gray-100 overflow-hidden min-h-0">
           {!fileUrl ? (
             <div className="flex items-center justify-center h-full text-center text-gray-500 p-8">
               <div>
                 <FileText className="w-16 h-16 mx-auto mb-4 text-gray-400" />
-                <p>No se ha proporcionado un archivo para previsualizar</p>
+                <p>No file available for preview</p>
               </div>
             </div>
           ) : (
             <>
-              {/* Indicador de carga */}
+              {/* Loading state */}
               {loading && (
                 <div className="absolute inset-0 flex items-center justify-center bg-white z-10">
                   <div className="text-center">
                     <Loader2 className="w-12 h-12 animate-spin text-blue-500 mx-auto mb-4" />
-                    <p className="text-gray-600">Cargando PDF...</p>
+                    <p className="text-gray-600">Loading PDF…</p>
                   </div>
                 </div>
               )}
 
-              {/*
-                Iframe para mostrar el PDF
-                El navegador maneja toda la visualización, zoom, scroll, etc.
-                Esto muestra el PDF exactamente como se vería si lo abrieras directamente
-              */}
               <iframe
                 src={fileUrl}
                 className="w-full h-full border-0"
-                title={`Vista previa de ${fileName}`}
+                title={`Preview of ${fileName}`}
                 onLoad={handleIframeLoad}
-                style={{ minHeight: '600px' }}
               />
             </>
           )}
         </div>
-
-        {/* Footer con información */}
-        <div className="px-6 py-3 border-t bg-gray-50 text-xs text-gray-500 text-center">
-          {fileName}
-        </div>
       </DialogContent>
     </Dialog>
-  )
+  );
 }

frontend/src/components/Sidebar.tsx

@@ -84,7 +84,7 @@ export function Sidebar({
   const handleCreateDataroom = async () => {
     const trimmed = newDataroomName.trim();
     if (!trimmed) {
-      setCreateError("El nombre es obligatorio");
+      setCreateError("Name is required");
       return;
     }
 
@@ -108,7 +108,7 @@ export function Sidebar({
       setCreateError(
         error instanceof Error
           ? error.message
-          : "No se pudo crear el dataroom. Inténtalo nuevamente.",
+          : "Could not create the dataroom. Please try again.",
       );
     } finally {
       setCreatingDataroom(false);
@@ -168,15 +168,15 @@ export function Sidebar({
     tema: string,
     e: React.MouseEvent<HTMLButtonElement>,
   ) => {
-    e.stopPropagation(); // Evitar que se seleccione el tema al hacer clic en el icono
+    e.stopPropagation(); // Prevent selecting the dataroom when clicking delete
 
     const confirmed = window.confirm(
-      `¿Estás seguro de que deseas eliminar el dataroom "${tema}"?\n\n` +
+      `Are you sure you want to delete the dataroom "${tema}"?\n\n` +
-        `Esto eliminará:\n` +
+        `This will remove:\n` +
-        `• El dataroom de la base de datos\n` +
+        `• The dataroom from the database\n` +
-        `• Todos los archivos del tema en Azure Blob Storage\n` +
+        `• All files stored for this topic in Azure Blob Storage\n` +
-        `• La colección "${tema}" en Qdrant (si existe)\n\n` +
+        `• The "${tema}" collection in Qdrant (if it exists)\n\n` +
-        `Esta acción no se puede deshacer.`,
+        `This action cannot be undone.`,
     );
 
     if (!confirmed) return;
@@ -191,10 +191,10 @@ export function Sidebar({
         console.error(`Error deleting dataroom "${tema}":`, error);
         // If dataroom deletion fails, fall back to legacy deletion
 
-        // Eliminar todos los archivos del tema en Azure Blob Storage
+        // Delete all topic files in Azure Blob Storage
         await api.deleteTema(tema);
 
-        // Intentar eliminar la colección en Qdrant (si existe)
+        // Attempt to delete the Qdrant collection (if it exists)
         try {
           const collectionExists = await api.checkCollectionExists(tema);
           if (collectionExists.exists) {
@@ -202,7 +202,7 @@ export function Sidebar({
           }
         } catch (collectionError) {
           console.warn(
-            `No se pudo eliminar la colección "${tema}" de Qdrant:`,
+            `Could not delete the "${tema}" collection from Qdrant:`,
             collectionError,
           );
         }
@@ -216,9 +216,9 @@ export function Sidebar({
         setSelectedTema(null);
       }
     } catch (error) {
-      console.error(`Error eliminando dataroom "${tema}":`, error);
+      console.error(`Error deleting dataroom "${tema}":`, error);
       alert(
-        `Error al eliminar el dataroom: ${error instanceof Error ? error.message : "Error desconocido"}`,
+        `Unable to delete dataroom: ${error instanceof Error ? error.message : "Unknown error"}`,
       );
     } finally {
       setDeletingTema(null);
@@ -251,9 +251,7 @@ export function Sidebar({
               className="text-slate-400 hover:text-slate-100"
               onClick={onToggleCollapse}
               disabled={disabled}
-              aria-label={
+              aria-label={collapsed ? "Expand sidebar" : "Collapse sidebar"}
-                collapsed ? "Expandir barra lateral" : "Contraer barra lateral"
-              }
             >
               {collapsed ? (
                 <ChevronRight className="h-4 w-4" />
@@ -273,16 +271,13 @@ export function Sidebar({
                 collapsed ? "justify-center" : "justify-between",
               )}
             >
-              <h2
+              {!collapsed && (
-                className={cn(
+                <h2 className="text-sm font-medium text-slate-300">
-                  "text-sm font-medium text-slate-300",
+                  Datarooms
-                  collapsed && "text-xs text-center",
-                )}
-              >
-                {collapsed ? "Rooms" : "Datarooms"}
                 </h2>
+              )}
               {renderWithTooltip(
-                "Crear dataroom",
+                "Create",
                 <Button
                   variant="ghost"
                   size="sm"
@@ -296,15 +291,15 @@ export function Sidebar({
                   disabled={disabled || creatingDataroom}
                 >
                   <Plus className="h-4 w-4" />
-                  {!collapsed && <span>Crear dataroom</span>}
+                  {!collapsed && <span>Create</span>}
                 </Button>,
               )}
             </div>
 
-            {/* Lista de temas */}
+            {/* Dataroom list */}
             {loading ? (
               <div className="text-sm text-slate-400 px-3 py-2 text-center">
-                {collapsed ? "..." : "Cargando..."}
+                {collapsed ? "..." : "Loading..."}
               </div>
             ) : Array.isArray(temas) && temas.length > 0 ? (
               temas.map((tema) => (
@@ -334,7 +329,7 @@ export function Sidebar({
                       onClick={(e) => handleDeleteTema(tema, e)}
                       disabled={deletingTema === tema || disabled}
                       className="absolute right-2 top-1/2 -translate-y-1/2 p-1.5 rounded hover:bg-red-500/20 opacity-0 group-hover:opacity-100 transition-opacity disabled:opacity-50"
-                      title="Eliminar dataroom y colección"
+                      title="Delete dataroom and collection"
                     >
                       <Trash2 className="h-4 w-4 text-red-400" />
                     </button>
@@ -344,8 +339,8 @@ export function Sidebar({
             ) : (
               <div className="text-sm text-slate-400 px-3 py-2 text-center">
                 {Array.isArray(temas) && temas.length === 0
-                  ? "No hay datarooms"
+                  ? "No datarooms found"
-                  : "Cargando datarooms..."}
+                  : "Loading datarooms..."}
               </div>
             )}
           </div>
@@ -360,7 +355,7 @@ export function Sidebar({
         >
           {onNavigateToSchemas &&
             renderWithTooltip(
-              "Gestionar Schemas",
+              "Manage schemas",
               <Button
                 variant="default"
                 size="sm"
@@ -373,12 +368,12 @@ export function Sidebar({
               >
                 <Database className={cn("h-4 w-4", !collapsed && "mr-2")} />
                 <span className={cn(collapsed && "sr-only")}>
-                  Gestionar Schemas
+                  Manage Schemas
                 </span>
               </Button>,
             )}
           {renderWithTooltip(
-            "Actualizar datarooms",
+            "Refresh datarooms",
             <Button
               variant="ghost"
               size="sm"
@@ -391,7 +386,7 @@ export function Sidebar({
             >
               <RefreshCcw className={cn("mr-2 h-4 w-4", collapsed && "mr-0")} />
               <span className={cn(collapsed && "sr-only")}>
-                Actualizar datarooms
+                Refresh datarooms
               </span>
             </Button>,
           )}
@@ -406,14 +401,14 @@ export function Sidebar({
           aria-describedby="create-dataroom-description"
         >
           <DialogHeader>
-            <DialogTitle>Crear dataroom</DialogTitle>
+            <DialogTitle>Create dataroom</DialogTitle>
             <DialogDescription id="create-dataroom-description">
-              Define un nombre único para organizar tus archivos.
+              Choose a unique name to organize your files.
             </DialogDescription>
           </DialogHeader>
           <div className="space-y-3">
             <div className="space-y-2">
-              <Label htmlFor="dataroom-name">Nombre del dataroom</Label>
+              <Label htmlFor="dataroom-name">Dataroom name</Label>
               <Input
                 id="dataroom-name"
                 value={newDataroomName}
@@ -423,7 +418,7 @@ export function Sidebar({
                     setCreateError(null);
                   }
                 }}
-                placeholder="Ej: normativa, contratos, fiscal..."
+                placeholder="e.g., policies, contracts, finance..."
                 autoFocus
               />
               {createError && (
@@ -437,13 +432,13 @@ export function Sidebar({
               onClick={() => handleCreateDialogOpenChange(false)}
               disabled={creatingDataroom}
             >
-              Cancelar
+              Cancel
             </Button>
             <Button
               onClick={handleCreateDataroom}
               disabled={creatingDataroom || newDataroomName.trim() === ""}
             >
-              {creatingDataroom ? "Creando..." : "Crear dataroom"}
+              {creatingDataroom ? "Creating…" : "Create"}
             </Button>
           </DialogFooter>
         </DialogContent>

frontend/src/components/WebSearchResults.tsx

@@ -0,0 +1,206 @@
+import React, { useState } from "react";
+import {
+  Globe,
+  ExternalLink,
+  Search,
+  ChevronDown,
+  ChevronRight,
+  Info,
+  Star,
+} from "lucide-react";
+import { cn } from "@/lib/utils";
+
+interface SearchResult {
+  title: string;
+  url: string;
+  content: string;
+  score?: number;
+}
+
+interface WebSearchData {
+  query: string;
+  results: SearchResult[];
+  summary: string;
+  total_results: number;
+}
+
+interface WebSearchResultsProps {
+  data: WebSearchData;
+}
+
+const getScoreColor = (score?: number) => {
+  if (!score) return "text-gray-500";
+  if (score >= 0.8) return "text-green-600";
+  if (score >= 0.6) return "text-yellow-600";
+  return "text-gray-500";
+};
+
+const getScoreStars = (score?: number) => {
+  if (!score) return 0;
+  return Math.round(score * 5);
+};
+
+const truncateContent = (content: string, maxLength: number = 200) => {
+  if (content.length <= maxLength) return content;
+  return content.slice(0, maxLength) + "...";
+};
+
+export const WebSearchResults: React.FC<WebSearchResultsProps> = ({ data }) => {
+  const [expandedResults, setExpandedResults] = useState<Set<number>>(new Set());
+  const [showAllResults, setShowAllResults] = useState(false);
+
+  const { query, results, summary, total_results } = data;
+
+  const toggleResult = (index: number) => {
+    const newExpanded = new Set(expandedResults);
+    if (newExpanded.has(index)) {
+      newExpanded.delete(index);
+    } else {
+      newExpanded.add(index);
+    }
+    setExpandedResults(newExpanded);
+  };
+
+  const visibleResults = showAllResults ? results : results.slice(0, 3);
+
+  return (
+    <div className="w-full bg-white border border-gray-200 rounded-lg shadow-sm overflow-hidden">
+      {/* Header */}
+      <div className="bg-gradient-to-r from-green-50 to-emerald-50 p-4">
+        <div className="flex items-center justify-between">
+          <div className="flex items-center gap-3">
+            <Globe className="w-6 h-6 text-green-600" />
+            <div>
+              <h3 className="font-semibold text-gray-900">Web Search Results</h3>
+              <div className="flex items-center gap-2 text-xs text-gray-600">
+                <Search className="w-3 h-3" />
+                <span>"{query}"</span>
+              </div>
+            </div>
+          </div>
+          <div className="text-right">
+            <div className="text-sm font-medium text-gray-900">{results.length}</div>
+            <div className="text-xs text-gray-600">
+              of {total_results} results
+            </div>
+          </div>
+        </div>
+      </div>
+
+      <div className="p-4 space-y-4">
+        {/* Summary */}
+        {summary && (
+          <div className="bg-blue-50 border border-blue-200 rounded-lg p-3">
+            <div className="flex items-start gap-2">
+              <Info className="w-4 h-4 text-blue-600 mt-0.5 flex-shrink-0" />
+              <div>
+                <h4 className="font-medium text-blue-900 text-sm mb-1">
+                  Summary
+                </h4>
+                <p className="text-sm text-blue-800">{summary}</p>
+              </div>
+            </div>
+          </div>
+        )}
+
+        {/* Search Results */}
+        <div className="space-y-3">
+          {visibleResults.map((result, index) => {
+            const isExpanded = expandedResults.has(index);
+            const stars = getScoreStars(result.score);
+
+            return (
+              <div key={index} className="border border-gray-200 rounded-lg overflow-hidden">
+                <div className="p-3">
+                  <div className="flex items-start justify-between mb-2">
+                    <div className="flex-1 min-w-0">
+                      <h4 className="font-medium text-gray-900 text-sm mb-1 line-clamp-2">
+                        {result.title}
+                      </h4>
+                      <div className="flex items-center gap-2">
+                        <a
+                          href={result.url}
+                          target="_blank"
+                          rel="noopener noreferrer"
+                          className="text-xs text-blue-600 hover:text-blue-800 flex items-center gap-1 truncate"
+                        >
+                          <ExternalLink className="w-3 h-3 flex-shrink-0" />
+                          {new URL(result.url).hostname}
+                        </a>
+                        {result.score && (
+                          <div className="flex items-center gap-1">
+                            <div className="flex">
+                              {[...Array(5)].map((_, i) => (
+                                <Star
+                                  key={i}
+                                  className={cn(
+                                    "w-3 h-3",
+                                    i < stars
+                                      ? "text-yellow-400 fill-current"
+                                      : "text-gray-300"
+                                  )}
+                                />
+                              ))}
+                            </div>
+                            <span className={cn("text-xs font-medium", getScoreColor(result.score))}>
+                              {Math.round((result.score || 0) * 100)}%
+                            </span>
+                          </div>
+                        )}
+                      </div>
+                    </div>
+                  </div>
+
+                  <div className="text-sm text-gray-700">
+                    {isExpanded ? result.content : truncateContent(result.content)}
+                  </div>
+
+                  {result.content.length > 200 && (
+                    <button
+                      onClick={() => toggleResult(index)}
+                      className="mt-2 flex items-center gap-1 text-xs text-blue-600 hover:text-blue-800 font-medium"
+                    >
+                      {isExpanded ? (
+                        <>
+                          <ChevronDown className="w-3 h-3" />
+                          Show less
+                        </>
+                      ) : (
+                        <>
+                          <ChevronRight className="w-3 h-3" />
+                          Read more
+                        </>
+                      )}
+                    </button>
+                  )}
+                </div>
+              </div>
+            );
+          })}
+        </div>
+
+        {/* Show More/Less Button */}
+        {results.length > 3 && (
+          <div className="text-center">
+            <button
+              onClick={() => setShowAllResults(!showAllResults)}
+              className="text-sm text-blue-600 hover:text-blue-800 font-medium px-4 py-2 rounded-lg hover:bg-blue-50 transition-colors"
+            >
+              {showAllResults
+                ? "Show fewer results"
+                : `Show ${results.length - 3} more results`}
+            </button>
+          </div>
+        )}
+
+        {/* No Results */}
+        {results.length === 0 && (
+          <div className="text-center py-6">
+            <Search className="w-8 h-8 text-gray-400 mx-auto mb-2" />
+            <p className="text-sm text-gray-600">No results found for "{query}"</p>
+          </div>
+        )}
+      </div>
+    </div>
+  );
+};

frontend/src/services/api.ts

@@ -39,6 +39,30 @@ interface DataroomsResponse {
   }>;
 }
 
+interface DataroomInfo {
+  name: string;
+  collection: string;
+  storage: string;
+  file_count: number;
+  total_size_bytes: number;
+  total_size_mb: number;
+  collection_exists: boolean;
+  vector_count: number | null;
+  collection_info: {
+    vectors_count: number;
+    indexed_vectors_count: number;
+    points_count: number;
+    segments_count: number;
+    status: string;
+  } | null;
+  file_types: Record<string, number>;
+  recent_files: Array<{
+    name: string;
+    size_mb: number;
+    last_modified: string;
+  }>;
+}
+
 interface CreateDataroomRequest {
   name: string;
   collection?: string;
@@ -100,6 +124,15 @@ export const api = {
     return response.json();
   },
 
+  // Obtener información detallada de un dataroom
+  getDataroomInfo: async (dataroomName: string): Promise<DataroomInfo> => {
+    const response = await fetch(
+      `${API_BASE_URL}/dataroom/${encodeURIComponent(dataroomName)}/info`,
+    );
+    if (!response.ok) throw new Error("Error fetching dataroom info");
+    return response.json();
+  },
+
   // Obtener archivos (todos o por tema)
   getFiles: async (tema?: string): Promise<FileListResponse> => {
     const url = tema

redis_data/appendonlydir/appendonly.aof.1.incr.aof

@@ -0,0 +1,791 @@
+*2
+$6
+SELECT
+$1
+0
+*24
+$14
+FT._CREATEIFNX
+$23
+extracted_doc:doc:index
+$2
+ON
+$4
+HASH
+$6
+PREFIX
+$1
+1
+$18
+extracted_doc:doc:
+$6
+SCHEMA
+$2
+pk
+$3
+TAG
+$9
+SEPARATOR
+$1
+|
+$9
+file_name
+$3
+TAG
+$9
+SEPARATOR
+$1
+|
+$4
+tema
+$3
+TAG
+$9
+SEPARATOR
+$1
+|
+$15
+collection_name
+$3
+TAG
+$9
+SEPARATOR
+$1
+|
+*3
+$3
+SET
+$28
+extracted_doc:doc:index:hash
+$40
+9de4cb60a645142de3d0f914909eb21259ea256c
+*12
+$14
+FT._CREATEIFNX
+$35
+:app.models.dataroom.DataRoom:index
+$2
+ON
+$4
+HASH
+$6
+PREFIX
+$1
+1
+$30
+:app.models.dataroom.DataRoom:
+$6
+SCHEMA
+$2
+pk
+$3
+TAG
+$9
+SEPARATOR
+$1
+|
+*3
+$3
+SET
+$40
+:app.models.dataroom.DataRoom:index:hash
+$40
+5ba27839f9c1b369df2b0734904dd56c02d2cea5
+*10
+$4
+HSET
+$56
+:app.models.dataroom.DataRoom:01K9J296J6FHT5AKG92D2VX9VW
+$2
+pk
+$26
+01K9J296J6FHT5AKG92D2VX9VW
+$4
+name
+$10
+ABBEY C.U.
+$10
+collection
+$0
+
+$7
+storage
+$0
+
+*2
+$3
+DEL
+$56
+:app.models.dataroom.DataRoom:01K9J296J6FHT5AKG92D2VX9VW
+*10
+$4
+HSET
+$56
+:app.models.dataroom.DataRoom:01K9J2J0A0HPWJQEY0PXFCVPW8
+$2
+pk
+$26
+01K9J2J0A0HPWJQEY0PXFCVPW8
+$4
+name
+$4
+ABBY
+$10
+collection
+$0
+
+$7
+storage
+$0
+
+*2
+$3
+DEL
+$56
+:app.models.dataroom.DataRoom:01K9J2J0A0HPWJQEY0PXFCVPW8
+*10
+$4
+HSET
+$56
+:app.models.dataroom.DataRoom:01K9J2R5ZGS96G60P0G80W248Z
+$2
+pk
+$26
+01K9J2R5ZGS96G60P0G80W248Z
+$4
+name
+$4
+ABBY
+$10
+collection
+$4
+abby
+$7
+storage
+$4
+abby
+*2
+$6
+SELECT
+$1
+0
+*14
+$4
+HSET
+$44
+extracted_doc:doc:01K9J44CG40W28WKNC6RD5QSGM
+$2
+pk
+$26
+01K9J44CG40W28WKNC6RD5QSGM
+$9
+file_name
+$21
+tax_year_2022_990.pdf
+$4
+tema
+$4
+ABBY
+$15
+collection_name
+$4
+ABBY
+$19
+extracted_data_json
+$5684
+{
+  "ein": "31-0329725",
+  "legal_name": "0220 ABBEY CREDIT UNION INC",
+  "phone_number": "(397) 898-7800",
+  "website_url": "www.abbeycu.com",
+  "return_type": "990",
+  "amended_return": "",
+  "group_exemption_number": "",
+  "subsection_code": "501(c)(14)",
+  "ruling_date": "",
+  "accounting_method": "Accrual",
+  "organization_type": "Corporation",
+  "year_of_formation": "1937",
+  "incorporation_state": "OH",
+  "total_revenue": 6146738,
+  "contributions_gifts_grants": 0,
+  "program_service_revenue": 5656278,
+  "membership_dues": 0,
+  "investment_income": 490460,
+  "gains_losses_sales_assets": 0,
+  "rental_income": 0,
+  "related_organizations_revenue": 0,
+  "gaming_revenue": 0,
+  "other_revenue": 0,
+  "government_grants": 0,
+  "foreign_contributions": 0,
+  "total_expenses": 5526970,
+  "program_services_expenses": 386454,
+  "management_general_expenses": 5140516,
+  "fundraising_expenses": 0,
+  "grants_us_organizations": 0,
+  "grants_us_individuals": 0,
+  "grants_foreign_organizations": 0,
+  "grants_foreign_individuals": 0,
+  "compensation_officers": 629393,
+  "compensation_other_staff": 1329887,
+  "payroll_taxes_benefits": 452511,
+  "professional_fees": 96471,
+  "office_occupancy_costs": 292244,
+  "information_technology_costs": 287125,
+  "travel_conference_expenses": 26662,
+  "depreciation_amortization": 236836,
+  "insurance": 39570,
+  "officers_list": [
+    "Lisa Burk, Chief Experience Officer, 40.00 hrs/wk, Officer, $73,286 compensation, $8,931 other compensation",
+    "Eric Stetzel, VP of Business Services, 40.00 hrs/wk, Key employee, $111,064 compensation, $15,979 other compensation",
+    "Dean Pielemeier, CEO, 40.00 hrs/wk, Officer, $212,994 compensation, $20,202 other compensation",
+    "Blanca Criner, Chief Marketing and Business Development Officer, 40.00 hrs/wk, Officer, $121,036 compensation, $4,540 other compensation",
+    "Teri Puthoff, CFO, 40.00 hrs/wk, Officer, $120,962 compensation, $15,754 other compensation",
+    "Michael Thein, Chairman, 1.00 hrs/wk, Individual trustee or director, $0 compensation",
+    "Latham Farley, Board Member, 1.00 hrs/wk, Individual trustee or director, $0 compensation",
+    "Steve Wilmoth, Treasurer, 1.00 hrs/wk, Individual trustee or director, $0 compensation",
+    "Julie Trick, Secretary, 1.00 hrs/wk, Individual trustee or director, $0 compensation",
+    "Michele Blake, Board Member, 1.00 hrs/wk, Individual trustee or director, $0 compensation",
+    "Cheryl Saunders, Board Member, 1.00 hrs/wk, Individual trustee or director, $0 compensation",
+    "Heather Scaggs-Richardson, Vice Chairman, 1.00 hrs/wk, Individual trustee or director, $0 compensation"
+  ],
+  "governing_body_size": 7,
+  "independent_members": 7,
+  "financial_statements_reviewed": "No",
+  "form_990_provided_to_governing_body": "Yes",
+  "conflict_of_interest_policy": "Yes",
+  "whistleblower_policy": "Yes",
+  "document_retention_policy": "Yes",
+  "ceo_compensation_review_process": "Compensation committee, independent compensation consultant, approval by the board or compensation committee",
+  "public_disclosure_practices": "Audited financial statements are handed out each year at the annual meeting. Other documents are not made available to the public unless requested.",
+  "program_accomplishments_list": [
+    "Our mission is to help our members improve their economic well being and quality of life by being competitive convenient and cutting edge.",
+    "Abbey provides checking savings accounts money market accounts certificates of deposit and IRAs. Abbey also offers free atm debit card services wire transfer services mobile banking and insurance related products.",
+    "Abbey provides personal vehicle mortgage and credit card loans to its members."
+  ],
+  "total_fundraising_event_revenue": 0,
+  "total_fundraising_event_expenses": 0,
+  "professional_fundraiser_fees": 0,
+  "number_of_employees": 50,
+  "number_of_volunteers": 7,
+  "occupancy_costs": 132399,
+  "fundraising_method_descriptions": "",
+  "joint_ventures_disregarded_entities": "",
+  "base_compensation": 546779,
+  "bonus": 34415,
+  "incentive": 59692,
+  "other_compensation": 24691,
+  "non_fixed_compensation": "",
+  "first_class_travel": "No",
+  "housing_allowance": "No",
+  "expense_account_usage": "No",
+  "supplemental_retirement": "No",
+  "lobbying_expenditures_direct": 0,
+  "lobbying_expenditures_grassroots": 0,
+  "election_501h_status": "",
+  "political_campaign_expenditures": 0,
+  "related_organizations_affiliates": "",
+  "investment_types": "Publicly traded securities",
+  "donor_restricted_endowment_values": 0,
+  "net_appreciation_depreciation": -2190939,
+  "related_organization_transactions": "",
+  "loans_to_from_related_parties": "Lynn Cook (Retired CEO) Life Insurance loan, $1,400,000; Dean Pielemeier (CEO) Life Insurance loan, $1,239,997",
+  "penalties_excise_taxes_reported": "No",
+  "unrelated_business_income_disclosure": "Yes",
+  "foreign_bank_account_reporting": "No",
+  "schedule_o_narrative_explanations": "Membership to Abbey is open to those who live work worship or attend school in Montgomery Miami Shelby Darke or Greene counties. Abbey CU is owned by the people who open a savings account at Abbey CU. Each member has one vote at the annual election of the Board of Directors. Decisions that require approval: Ohio Division of Financial Institutions Pursuant to regulation. CFO prepares the form and CEO reviews before submitting. Approval of Compensation Committee and Board of Directors for CEO compensation. CEO establishes compensation for Key Employees. Audited financial statements are handed out each year at the annual meeting. Other documents are not made available to the public unless requested."
+}
+$20
+extraction_timestamp
+$26
+2025-11-08T16:17:31.137820
+*2
+$6
+SELECT
+$1
+0
+*14
+$4
+HSET
+$44
+extracted_doc:doc:01K9MH40SHHNWXNGAP4MDX6HZF
+$2
+pk
+$26
+01K9MH40SHHNWXNGAP4MDX6HZF
+$9
+file_name
+$21
+tax_year_2019_990.pdf
+$4
+tema
+$4
+ABBY
+$15
+collection_name
+$4
+ABBY
+$19
+extracted_data_json
+$4954
+{
+  "ein": "31-0329725",
+  "legal_name": "Abbey Credit Union Inc",
+  "phone_number": "937-898-7800",
+  "website_url": "www.abbeycu.com",
+  "return_type": "990",
+  "amended_return": "",
+  "group_exemption_number": "",
+  "subsection_code": "501(c)(14)",
+  "ruling_date": "",
+  "accounting_method": "Accrual",
+  "organization_type": "Corporation",
+  "year_of_formation": "1937",
+  "incorporation_state": "OH",
+  "total_revenue": 4659611,
+  "contributions_gifts_grants": 0,
+  "program_service_revenue": 4114069,
+  "membership_dues": 0,
+  "investment_income": 545542,
+  "gains_losses_sales_assets": 0,
+  "rental_income": 6725,
+  "related_organizations_revenue": 0,
+  "gaming_revenue": 0,
+  "other_revenue": 0,
+  "government_grants": 0,
+  "foreign_contributions": 0,
+  "total_expenses": 4159254,
+  "program_services_expenses": 0,
+  "management_general_expenses": 0,
+  "fundraising_expenses": 0,
+  "grants_us_organizations": 0,
+  "grants_us_individuals": 0,
+  "grants_foreign_organizations": 0,
+  "grants_foreign_individuals": 0,
+  "compensation_officers": 338208,
+  "compensation_other_staff": 1018956,
+  "payroll_taxes_benefits": 309032,
+  "professional_fees": 34366,
+  "office_occupancy_costs": 0,
+  "information_technology_costs": 0,
+  "travel_conference_expenses": 0,
+  "depreciation_amortization": 0,
+  "insurance": 0,
+  "officers_list": [
+    "Michael Thein, Chairman",
+    "Nancy Wood, Vice Chairman",
+    "Steve Wilmoth, Treasurer",
+    "Julie Trick, Secretary",
+    "Michele Blake, Board Member",
+    "Cheryl Saunders, Board Member",
+    "Heather Scaggs-Richardson, Board Member",
+    "Dean Pielemeier, CEO",
+    "Teri Puthoff, VP of Finance",
+    "Blanca Ortiz, VP of Business Development"
+  ],
+  "governing_body_size": 7,
+  "independent_members": 7,
+  "financial_statements_reviewed": "Yes",
+  "form_990_provided_to_governing_body": "Yes",
+  "conflict_of_interest_policy": "Yes",
+  "whistleblower_policy": "Yes",
+  "document_retention_policy": "Yes",
+  "ceo_compensation_review_process": "Compensation for Dean Pielemeier, CEO is reviewed by the Personnel Committee of the board of directors and is acted upon by the board of directors per committee recommendation Compensation for officers is determined by Dean Pielemeier, CEO",
+  "public_disclosure_practices": "Documents are not made available to the public",
+  "program_accomplishments_list": [
+    "Our mission is to help our members improve their economic well-being and quality of life by being competitive, convenient, and cutting edge",
+    "LENDING - ABBEY PROVIDES PERSONAL, VEHICLE, MORTGAGE, AND CREDIT CARD LOANS TO ITS MEMBERS.",
+    "FINANCIAL SERVICES - ABBEY PROVIDES CHECKING ACCOUNTS, SAVINGS ACCOUNTS, MONEY MARKET ACCOUNTS"
+  ],
+  "total_fundraising_event_revenue": 0,
+  "total_fundraising_event_expenses": 0,
+  "professional_fundraiser_fees": 0,
+  "number_of_employees": 36,
+  "number_of_volunteers": 7,
+  "occupancy_costs": 0,
+  "fundraising_method_descriptions": "",
+  "joint_ventures_disregarded_entities": "",
+  "base_compensation": 154796,
+  "bonus": 6409,
+  "incentive": 0,
+  "other_compensation": 4642,
+  "non_fixed_compensation": "",
+  "first_class_travel": "",
+  "housing_allowance": "",
+  "expense_account_usage": "",
+  "supplemental_retirement": "Yes",
+  "lobbying_expenditures_direct": 0,
+  "lobbying_expenditures_grassroots": 0,
+  "election_501h_status": "",
+  "political_campaign_expenditures": 0,
+  "related_organizations_affiliates": "",
+  "investment_types": "",
+  "donor_restricted_endowment_values": 0,
+  "net_appreciation_depreciation": 281072,
+  "related_organization_transactions": "",
+  "loans_to_from_related_parties": "LYNN COOK, PRIOR CEO, LIFE INSURAN, Loan from organization, $140,000 balance due",
+  "penalties_excise_taxes_reported": "",
+  "unrelated_business_income_disclosure": "Yes",
+  "foreign_bank_account_reporting": "No",
+  "schedule_o_narrative_explanations": "Our mission is to help our members improve their economic well-being and quality of life by being competitive, convenient, and cutting edge. Membership to Abbey is open to those who live, work, or worship, or attend school in Montgomery, Miami, Shelby, Darke, or Greene counties. Abbey Credit Union is owned by these people that open an account at Abbey CU. Yes - Each member gets one vote at the annual election. Yes - Pursuant to the regulation of the Ohio Division of Credit Unions. The VP of Finance prepared the Form 990 based on financial records Financial statements are audited annually The CEO reviews the returns prior to filing. The Credit Union has a written conflict of interest policy that states that board members are responsible for disclosing possible conflicts of interest as they arise This is reviewed annually. Compensation for Dean Pielemeier, CEO is reviewed by the Personnel Committee of the board of directors and is acted upon by the board of directors per committee recommendation Compensation for officers is determined by Dean Pielemeier, CEO."
+}
+$20
+extraction_timestamp
+$26
+2025-11-09T14:42:59.494840
+*10
+$4
+HSET
+$56
+:app.models.dataroom.DataRoom:01K9MNBVRY6KVF4XCK70JF70T0
+$2
+pk
+$26
+01K9MNBVRY6KVF4XCK70JF70T0
+$4
+name
+$5
+ABBEY
+$10
+collection
+$5
+abbey
+$7
+storage
+$5
+abbey
+*14
+$4
+HSET
+$44
+extracted_doc:doc:01K9MNMNYH1MYP0TQ5SWZRS5BG
+$2
+pk
+$26
+01K9MNMNYH1MYP0TQ5SWZRS5BG
+$9
+file_name
+$21
+tax_year_2022_990.pdf
+$4
+tema
+$5
+ABBEY
+$15
+collection_name
+$5
+ABBEY
+$19
+extracted_data_json
+$5525
+{
+  "ein": "31-0329725",
+  "calendar_year": 2022,
+  "legal_name": "0220 ABBEY CREDIT UNION INC",
+  "phone_number": "(397) 898-7800",
+  "website_url": "www.abbeycu.com",
+  "return_type": "990",
+  "amended_return": "",
+  "group_exemption_number": "",
+  "subsection_code": "501(c)(14)",
+  "ruling_date": "",
+  "accounting_method": "Accrual",
+  "organization_type": "Corporation",
+  "year_of_formation": "1937",
+  "incorporation_state": "OH",
+  "total_revenue": 6146738,
+  "contributions_gifts_grants": 0,
+  "program_service_revenue": 5656278,
+  "membership_dues": 0,
+  "investment_income": 490460,
+  "gains_losses_sales_assets": 0,
+  "rental_income": 0,
+  "related_organizations_revenue": 0,
+  "gaming_revenue": 0,
+  "other_revenue": 0,
+  "government_grants": 0,
+  "foreign_contributions": 0,
+  "total_expenses": 5526970,
+  "program_services_expenses": 386454,
+  "management_general_expenses": 5140516,
+  "fundraising_expenses": 0,
+  "grants_us_organizations": 0,
+  "grants_us_individuals": 0,
+  "grants_foreign_organizations": 0,
+  "grants_foreign_individuals": 0,
+  "compensation_officers": 629393,
+  "compensation_other_staff": 1329887,
+  "payroll_taxes_benefits": 452511,
+  "professional_fees": 96462,
+  "office_occupancy_costs": 292244,
+  "information_technology_costs": 287125,
+  "travel_conference_expenses": 26662,
+  "depreciation_amortization": 236836,
+  "insurance": 39570,
+  "officers_list": [
+    "Lisa Burk, Chief Experience Officer, 40.00 hrs/wk, Officer, $73,286 compensation, $8,931 other compensation",
+    "Eric Stetzel, VP of Business Services, 40.00 hrs/wk, Key employee, $111,064 compensation, $15,979 other compensation",
+    "Dean Pielemeier, CEO, 40.00 hrs/wk, Officer, $212,994 compensation, $20,202 other compensation",
+    "Blanca Criner, Chief Marketing and Business Development Officer, 40.00 hrs/wk, Officer, $121,036 compensation, $4,540 other compensation",
+    "Teri Puthoff, CFO, 40.00 hrs/wk, Officer, $120,962 compensation, $15,754 other compensation",
+    "Michael Thein, Chairman, 1.00 hrs/wk, Individual trustee or director, $0 compensation",
+    "Latham Farley, Board Member, 1.00 hrs/wk, Individual trustee or director, $0 compensation",
+    "Steve Wilmoth, Treasurer, 1.00 hrs/wk, Individual trustee or director, $0 compensation",
+    "Julie Trick, Secretary, 1.00 hrs/wk, Individual trustee or director, $0 compensation",
+    "Michele Blake, Board Member, 1.00 hrs/wk, Individual trustee or director, $0 compensation",
+    "Cheryl Saunders, Board Member, 1.00 hrs/wk, Individual trustee or director, $0 compensation",
+    "Heather Scaggs-Richardson, Vice Chairman, 1.00 hrs/wk, Individual trustee or director, $0 compensation"
+  ],
+  "governing_body_size": 7,
+  "independent_members": 7,
+  "financial_statements_reviewed": "No",
+  "form_990_provided_to_governing_body": "Yes",
+  "conflict_of_interest_policy": "Yes",
+  "whistleblower_policy": "Yes",
+  "document_retention_policy": "Yes",
+  "ceo_compensation_review_process": "Approval of Compensation Committee and Board of Directors",
+  "public_disclosure_practices": "Audited financial statements are handed out each year at the annual meeting. Other documents are not made available to the public unless requested.",
+  "program_accomplishments_list": [
+    "Abbey provides checking savings accounts money market accounts certificates of deposit and IRAs. Abbey also offers free atm debit card services wire transfer services mobile banking and insurance related products.",
+    "Abbey provides personal vehicle mortgage and credit card loans to its members."
+  ],
+  "total_fundraising_event_revenue": 0,
+  "total_fundraising_event_expenses": 0,
+  "professional_fundraiser_fees": 0,
+  "number_of_employees": 50,
+  "number_of_volunteers": 7,
+  "occupancy_costs": 132399,
+  "fundraising_method_descriptions": "",
+  "joint_ventures_disregarded_entities": "",
+  "base_compensation": 544779,
+  "bonus": 34315,
+  "incentive": 59692,
+  "other_compensation": 0,
+  "non_fixed_compensation": "",
+  "first_class_travel": "No",
+  "housing_allowance": "No",
+  "expense_account_usage": "No",
+  "supplemental_retirement": "No",
+  "lobbying_expenditures_direct": 0,
+  "lobbying_expenditures_grassroots": 0,
+  "election_501h_status": "",
+  "political_campaign_expenditures": 0,
+  "related_organizations_affiliates": "",
+  "investment_types": "Publicly traded securities",
+  "donor_restricted_endowment_values": 0,
+  "net_appreciation_depreciation": -2190939,
+  "related_organization_transactions": "",
+  "loans_to_from_related_parties": "Lynn Cook (Retired CEO) and Dean Pielemeier (CEO) received life insurance loans from the organization, each with a balance due of $1,400,000 and $1,239,997 respectively.",
+  "penalties_excise_taxes_reported": "No",
+  "unrelated_business_income_disclosure": "Yes",
+  "foreign_bank_account_reporting": "No",
+  "schedule_o_narrative_explanations": "Membership to Abbey is open to those who live work worship or attend school in Montgomery Miami Shelby Darke or Greene counties. Abbey CU is owned by the people who open a savings account at Abbey CU. Each member has one vote at the annual election of the Board of Directors. Decisions that require their approval: Ohio Division of Financial Institutions Pursuant to regulation. CFO prepares the form and CEO reviews before submitting. Approval of Compensation Committee and Board of Directors for CEO compensation. Audited financial statements are handed out each year at the annual meeting. Other documents are not made available to the public unless requested."
+}
+$20
+extraction_timestamp
+$26
+2025-11-09T16:01:59.759953
+*14
+$4
+HSET
+$44
+extracted_doc:doc:01K9MP8DBNWMRH9A966RRJ4E7V
+$2
+pk
+$26
+01K9MP8DBNWMRH9A966RRJ4E7V
+$9
+file_name
+$21
+tax_year_2019_990.pdf
+$4
+tema
+$5
+ABBEY
+$15
+collection_name
+$5
+ABBEY
+$19
+extracted_data_json
+$4959
+{
+  "ein": "31-0329725",
+  "calendar_year": 2019,
+  "legal_name": "Abbey Credit Union Inc",
+  "phone_number": "937-898-7800",
+  "website_url": "www.abbeycu.com",
+  "return_type": "990",
+  "amended_return": "",
+  "group_exemption_number": "",
+  "subsection_code": "501(c)(14)",
+  "ruling_date": "",
+  "accounting_method": "Accrual",
+  "organization_type": "Corporation",
+  "year_of_formation": "1937",
+  "incorporation_state": "OH",
+  "total_revenue": 4659611,
+  "contributions_gifts_grants": 0,
+  "program_service_revenue": 4114069,
+  "membership_dues": 0,
+  "investment_income": 545542,
+  "gains_losses_sales_assets": 0,
+  "rental_income": 6725,
+  "related_organizations_revenue": 0,
+  "gaming_revenue": 0,
+  "other_revenue": 0,
+  "government_grants": 0,
+  "foreign_contributions": 0,
+  "total_expenses": 4159254,
+  "program_services_expenses": 0,
+  "management_general_expenses": 0,
+  "fundraising_expenses": 0,
+  "grants_us_organizations": 0,
+  "grants_us_individuals": 0,
+  "grants_foreign_organizations": 0,
+  "grants_foreign_individuals": 0,
+  "compensation_officers": 338208,
+  "compensation_other_staff": 1018956,
+  "payroll_taxes_benefits": 273057,
+  "professional_fees": 34366,
+  "office_occupancy_costs": 200152,
+  "information_technology_costs": 0,
+  "travel_conference_expenses": 46563,
+  "depreciation_amortization": 0,
+  "insurance": 0,
+  "officers_list": {
+    "value": [
+      "Michael Thein, Chairman",
+      "Nancy Wood, Vice Chairman",
+      "Steve Wilmoth, Treasurer",
+      "Julie Trick, Secretary",
+      "Michele Blake, Board Member",
+      "Cheryl Saunders, Board Member",
+      "Heather Scaggs-Richardson, Board Member",
+      "Dean Pielemeier, CEO",
+      "Teri Puthoff, VP of Finance",
+      "Blanca Ortiz, VP of Business Development"
+    ],
+    "chunk_references": []
+  },
+  "governing_body_size": 7,
+  "independent_members": 7,
+  "financial_statements_reviewed": "Yes",
+  "form_990_provided_to_governing_body": "Yes",
+  "conflict_of_interest_policy": "Yes",
+  "whistleblower_policy": "Yes",
+  "document_retention_policy": "Yes",
+  "ceo_compensation_review_process": "Compensation for Dean Pielemeier, CEO is reviewed by the Personnel Committee of the board of directors and is acted upon by the board of directors per committee recommendation.",
+  "public_disclosure_practices": "Documents are not made available to the public",
+  "program_accomplishments_list": {
+    "value": [
+      "LENDING - ABBEY PROVIDES PERSONAL, VEHICLE, MORTGAGE, AND CREDIT CARD LOANS TO ITS MEMBERS.",
+      "FINANCIAL SERVICES - ABBEY PROVIDES CHECKING ACCOUNTS, SAVINGS ACCOUNTS, MONEY MARKET ACCOUNTS, ..."
+    ],
+    "chunk_references": []
+  },
+  "total_fundraising_event_revenue": 0,
+  "total_fundraising_event_expenses": 0,
+  "professional_fundraiser_fees": 0,
+  "number_of_employees": 36,
+  "number_of_volunteers": 7,
+  "occupancy_costs": 87912,
+  "fundraising_method_descriptions": "",
+  "joint_ventures_disregarded_entities": "",
+  "base_compensation": 154796,
+  "bonus": 6409,
+  "incentive": 0,
+  "other_compensation": 4642,
+  "non_fixed_compensation": "",
+  "first_class_travel": "",
+  "housing_allowance": "",
+  "expense_account_usage": "",
+  "supplemental_retirement": "Yes",
+  "lobbying_expenditures_direct": 0,
+  "lobbying_expenditures_grassroots": 0,
+  "election_501h_status": "",
+  "political_campaign_expenditures": 0,
+  "related_organizations_affiliates": "",
+  "investment_types": "",
+  "donor_restricted_endowment_values": 0,
+  "net_appreciation_depreciation": 281072,
+  "related_organization_transactions": "",
+  "loans_to_from_related_parties": "Loans to prior CEO Lynn Cook for life insurance, $1,400,000 total outstanding.",
+  "penalties_excise_taxes_reported": "",
+  "unrelated_business_income_disclosure": "Yes",
+  "foreign_bank_account_reporting": "No",
+  "schedule_o_narrative_explanations": "Our mission is to help our members improve their economic well-being and quality of life by being competitive, convenient, and cutting edge. Membership to Abbey is open to those who live, work, or worship, or attend school in Montgomery, Miami, Shelby, Darke, or Greene counties. Abbey Credit Union is owned by these people that open an account at Abbey CU. Yes - Each member gets one vote at the annual election. Yes - Pursuant to the regulation of the Ohio Division of Credit Unions. The VP of Finance prepared the Form 990 based on financial records. Financial statements are audited annually. The CEO reviews the returns prior to filing. The Credit Union has a written conflict of interest policy that states that board members are responsible for disclosing possible conflicts of interest as they arise. This is reviewed annually. Compensation for Dean Pielemeier, CEO is reviewed by the Personnel Committee of the board of directors and is acted upon by the board of directors per committee recommendation. Compensation for officers is determined by Dean Pielemeier, CEO. Documents are not made available to the public."
+}
+$20
+extraction_timestamp
+$26
+2025-11-09T16:12:46.318554
+*2
+$3
+DEL
+$56
+:app.models.dataroom.DataRoom:01K9J2R5ZGS96G60P0G80W248Z
+*10
+$4
+HSET
+$56
+:app.models.dataroom.DataRoom:01K9MSH70T90HY2VB6DPJVQZFQ
+$2
+pk
+$26
+01K9MSH70T90HY2VB6DPJVQZFQ
+$4
+name
+$9
+NEW HEART
+$10
+collection
+$9
+new_heart
+$7
+storage
+$9
+new_heart
+*10
+$4
+HSET
+$56
+:app.models.dataroom.DataRoom:01K9MSJ57MS3DZFBG5TQXBDD6W
+$2
+pk
+$26
+01K9MSJ57MS3DZFBG5TQXBDD6W
+$4
+name
+$5
+OHANA
+$10
+collection
+$5
+ohana
+$7
+storage
+$5
+ohana
+*10
+$4
+HSET
+$56
+:app.models.dataroom.DataRoom:01K9MSJJTH48BR7KQ27PXB2C3S
+$2
+pk
+$26
+01K9MSJJTH48BR7KQ27PXB2C3S
+$4
+name
+$5
+SOSFS
+$10
+collection
+$5
+sosfs
+$7
+storage
+$5
+sosfs