add healthcheck to remaining apps

apps/egresos/api/agent/__init__.py

@@ -0,0 +1,3 @@
+from .main import Agent
+
+__all__ = ["Agent"]

apps/egresos/api/agent/main.py

@@ -0,0 +1,108 @@
+from pathlib import Path
+from typing import Any
+
+from langchain_core.messages import AIMessageChunk
+from pydantic import BaseModel, Field
+from langchain_azure_ai.chat_models import AzureAIChatCompletionsModel
+from langchain_azure_ai.embeddings import AzureAIEmbeddingsModel
+
+from banortegpt.vector.qdrant import AsyncQdrant
+
+from api import context
+from api.config import config
+
+parent = Path(__file__).parent
+SYSTEM_PROMPT = (parent / "system_prompt.md").read_text()
+
+AZURE_AI_URI = "https://eastus2.api.cognitive.microsoft.com"
+
+class get_information(BaseModel):
+    """Search a private repository for information."""
+
+    question: str = Field(..., description="The user question")
+
+class Agent:
+    system_prompt = SYSTEM_PROMPT
+    generation_config = {
+        "temperature": config.model_temperature,
+    }
+    embedding_model = config.embedding_model
+    message_limit = config.message_limit
+    index = config.vector_index
+    limit = config.search_limit
+
+    search = AsyncQdrant.from_config(config)
+    llm = AzureAIChatCompletionsModel(
+        endpoint=f"{AZURE_AI_URI}/openai/deployments/{config.model}",
+        credential=config.openai_api_key,
+    ).bind_tools([get_information])
+    embedder = AzureAIEmbeddingsModel(
+        endpoint=f"{AZURE_AI_URI}/openai/deployments/{config.embedding_model}",
+        credential=config.openai_api_key,
+    )
+  
+    def __init__(self) -> None:
+        self.tool_map = {
+            "get_information": self.get_information
+        }
+
+    def build_response(self, payloads, fallback):
+        template = "<FAQ {index}>\n\n{content}\n\n</FAQ {index}>"
+        
+        filled_templates = [
+            template.format(index=idx, content=payload["content"])
+            for idx, payload in enumerate(payloads)
+        ]
+        filled_templates.append(f"<FALLBACK>\n{fallback}\n</FALLBACK>")
+        
+        return "\n".join(filled_templates)
+
+    async def get_information(self, question: str):
+        embedding = await self.embedder.aembed_query(question)
+        
+        payloads = await self.search.semantic_search(
+            embedding=embedding,
+            collection=self.index,
+            limit=self.limit,
+        )
+        
+        fallback_messages = {}
+        images = []
+        for idx, payload in enumerate(payloads):
+            fallback_message = payload.get("fallback_message", "None")
+            fallback_messages[fallback_message] = fallback_messages.get(fallback_message, 0) + 1
+
+            # Solo extraer imágenes del primer payload
+            if idx == 0 and "images" in payload:    
+                images.extend(payload["images"])
+
+        fallback = max(fallback_messages, key=fallback_messages.get)  # type: ignore
+        
+        response = self.build_response(payloads, fallback)
+        return str(response), images[:3]  # Limitar a 3 imágenes máximo
+
+    def _generation_config_overwrite(self, overwrites: dict | None) -> dict[str, Any]:
+        if not overwrites:
+            return self.generation_config.copy()
+        return {**self.generation_config, **overwrites}
+
+    async def stream(self, history, overwrites: dict | None = None):
+        generation_config = self._generation_config_overwrite(overwrites)
+
+        async for delta in self.llm.astream(input=history, **generation_config):
+            assert isinstance(delta, AIMessageChunk)
+            if call := delta.tool_call_chunks:
+                if tool_id := call[0].get("id"):
+                    context.tool_id.set(tool_id)
+                if name := call[0].get("name"):
+                    context.tool_name.set(name)
+                if args := call[0].get("args"):
+                    context.tool_buffer.set(context.tool_buffer.get() + args)
+            elif delta.content:
+                assert isinstance(delta.content, str)
+                context.buffer.set(context.buffer.get() + delta.content)
+                yield delta.content
+
+    async def generate(self, history, overwrites: dict | None = None):
+        generation_config = self._generation_config_overwrite(overwrites)
+        return await self.llm.ainvoke(input=history, **generation_config)

apps/egresos/api/agent/system_prompt.md

@@ -0,0 +1,49 @@
+🧠 Asistente Experto en la Política de Gastos de Viaje — Banorte
+🎯 Rol del Asistente:
+Especialista normativo encargado de responder exclusivamente con base en la Política Oficial de Gastos de Viaje de Banorte, garantizando respuestas profesionales, claras y verificables.
+
+✅ Misión Principal:
+Brindar respuestas 100% alineadas con la política vigente de gastos de viaje de Banorte, cumpliendo con los siguientes principios:
+
+⚙️ Reglas de Respuesta (Obligatorias):
+📥 Consulta siempre con get_information:
+Toda respuesta debe obtenerse únicamente a través de la herramienta get_information(question), que consulta la base de datos vectorial autorizada.
+
+Esta herramienta tambien cuenta con la constancia de sitaicion fiscal de banorte en un url 
+
+No es obligatorio que el usuario especifique estrictamente su puesto para realizar la consulta.
+
+Si el usuario sí indica un puesto, la respuesta debe forzarse a ese puesto y aplicarse la información correspondiente.
+
+En caso de que no exista información para el puesto indicado, se debe responder con la respuesta general disponible en la base de conocimiento.
+    
+❗ Nunca inventar ni responder sin antes consultar esta fuente.
+
+Si la herramienta no devuelve información relevante, indicar que la política no contempla esa situación.
+
+📚 Fuente única y oficial:
+Las respuestas deben estar basadas únicamente en la política oficial de Banorte.
+
+❌ Prohibido usar Google, foros, suposiciones o contenido externo.
+
+✅ Si get_information devuelve un enlace oficial o documento, debe incluirse con el ícono:
+🔗 [Ver política oficial].
+
+📐 Formato estructurado y profesional:
+Utilizar un formato claro y fácil de leer:
+• Viñetas para listar pasos, excepciones o montos autorizados
+• Negritas para resaltar conceptos clave
+• Separación clara entre secciones
+
+🔒 Cero invención o interpretación libre:
+Si una pregunta no está contemplada en la política, responder claramente:
+
+❗ La política oficial no proporciona lineamientos específicos sobre este caso.
+
+💼 Tono ejecutivo y directo:
+
+Profesional y objetivo
+
+Sin tecnicismos innecesarios
+
+Redacción breve, clara y enfocada en lo esencial 

apps/egresos/api/config.py

@@ -0,0 +1,59 @@
+from hvac import Client
+from pydantic import Field
+from pydantic_settings import BaseSettings
+
+client = Client(url="https://vault.ia-innovacion.work")
+
+if not client.is_authenticated():
+    raise Exception("Vault authentication failed")
+
+secret_map = client.secrets.kv.v2.read_secret_version(
+    path="banortegpt", mount_point="secret"
+)["data"]["data"]
+
+class Settings(BaseSettings):
+    """
+    Esta clase obtiene sus valores de variables de ambiente.
+    Si no estan en el ambiente, los jala de nuestra Vault.
+    """
+
+    # Config
+    model: str = "gpt-4o"
+    model_temperature: int = 0
+    message_limit: int = 10
+    host: str = "0.0.0.0"
+    port: int = 8000
+    vector_index: str = "chat-egresos-3"
+    search_limit: int = 3
+    embedding_model: str = "text-embedding-3-large"
+
+    # API Keys
+    azure_endpoint: str = Field(default_factory=lambda: secret_map["azure_endpoint"])
+    openai_api_key: str = Field(default_factory=lambda: secret_map["openai_api_key"])
+    openai_api_version: str = Field(
+        default_factory=lambda: secret_map["openai_api_version"]
+    )
+    mongodb_url: str = Field(
+        default_factory=lambda: secret_map["cosmosdb_connection_string"]
+    )
+
+    qdrant_url: str = Field(default_factory=lambda: secret_map["qdrant_api_url"])
+    qdrant_api_key: str | None = Field(
+        default_factory=lambda: secret_map["qdrant_api_key"]
+    )
+
+    async def init_mongo_db(self):
+        """Este helper inicia la conexion enter el MongoDB ORM y nuestra instancia"""
+
+        from beanie import init_beanie
+        from motor.motor_asyncio import AsyncIOMotorClient
+
+        from banortegpt.database.mongo_memory.models import Conversation
+
+        await init_beanie(
+            database=AsyncIOMotorClient(self.mongodb_url).banortegptdos,
+            document_models=[Conversation],
+        )
+
+
+config = Settings()

apps/egresos/api/context.py

@@ -0,0 +1,6 @@
+from contextvars import ContextVar
+
+buffer: ContextVar[str] = ContextVar("buffer", default="")
+tool_buffer: ContextVar[str] = ContextVar("tool_buffer", default="")
+tool_id: ContextVar[str | None] = ContextVar("tool_id", default=None)
+tool_name: ContextVar[str | None] = ContextVar("tool_name", default=None)

apps/egresos/api/server.py

@@ -0,0 +1,116 @@
+import time
+import uuid
+from contextlib import asynccontextmanager
+
+from fastapi import FastAPI
+from fastapi.responses import StreamingResponse
+from langfuse import Langfuse
+from pydantic import BaseModel
+
+from api import services
+from api.agent import Agent
+from api.config import config
+
+# Configurar Langfuse
+langfuse = Langfuse(
+    public_key="pk-lf-49cb04b3-0c7d-475b-8105-ad8b8749ecdd",
+    secret_key="sk-lf-e02fa322-c709-4d80-bef2-9cb279846a0c",
+    host="https://ailogger.azurewebsites.net",
+)
+
+
+@asynccontextmanager
+async def lifespan(_: FastAPI):
+    await config.init_mongo_db()
+    yield
+
+
+app = FastAPI(lifespan=lifespan)
+agent = Agent()
+
+
+@app.post("/api/v1/conversation")
+async def create_conversation():
+    conversation_id = uuid.uuid4()
+    await services.create_conversation(conversation_id, agent.system_prompt)
+    return {"conversation_id": conversation_id}
+
+
+class Message(BaseModel):
+    conversation_id: uuid.UUID
+    prompt: str
+
+
+@app.post("/api/v1/message")
+async def send(message: Message):
+    # Crear trace principal
+    trace = langfuse.trace(
+        name="chat_message",
+        session_id=str(message.conversation_id),
+        input={
+            "prompt": message.prompt,
+            "conversation_id": str(message.conversation_id),
+        },
+    )
+
+    def b64_sse(func):
+        async def wrapper(*args, **kwargs):
+            response_parts = []
+            start_time = time.time()
+
+            async for chunk in func(*args, **kwargs):
+                if chunk.type == "text" and chunk.content:
+                    response_parts.append(str(chunk.content))
+
+                content = chunk.model_dump_json()
+                data = f"data: {content}\n\n"
+                yield data
+
+            end_time = time.time()
+            latency_ms = round((end_time - start_time) * 1000)
+            full_response = "".join(response_parts)
+
+            input_tokens = len(message.prompt.split()) * 1.3
+            output_tokens = len(full_response.split()) * 1.3
+            total_tokens = int(input_tokens + output_tokens)
+
+            cost_per_1k_input = 0.03
+            cost_per_1k_output = 0.06
+            total_cost = (input_tokens / 1000 * cost_per_1k_input) + (
+                output_tokens / 1000 * cost_per_1k_output
+            )
+
+            trace.update(
+                output={"response": full_response},
+                usage={
+                    "input": int(input_tokens),
+                    "output": int(output_tokens),
+                    "total": total_tokens,
+                    "unit": "TOKENS",
+                },
+            )
+
+            langfuse.score(
+                trace_id=trace.id,
+                name="latency",
+                value=latency_ms,
+                comment=f"Response time: {latency_ms}ms",
+            )
+
+            langfuse.score(
+                trace_id=trace.id,
+                name="cost",
+                value=round(total_cost, 4),
+                comment=f"Estimated cost: ${round(total_cost, 4)}",
+            )
+
+        return wrapper
+
+    sse_stream = b64_sse(services.stream)
+    generator = sse_stream(agent, message.prompt, message.conversation_id)
+    return StreamingResponse(generator, media_type="text/event-stream")
+
+
+@app.get("/")
+async def health():
+    return {"status": "ok"}

apps/egresos/api/services/__init__.py

@@ -0,0 +1,8 @@
+from banortegpt.database.mongo_memory.crud import create_conversation
+
+from .stream_response import stream
+
+__all__ = [
+    "stream",
+    "create_conversation",
+]

apps/egresos/api/services/stream_response.py

@@ -0,0 +1,86 @@
+import json
+from enum import StrEnum
+from typing import TypeAlias
+from uuid import UUID
+
+from pydantic import BaseModel
+
+import api.context as ctx
+from api.agent import Agent
+from banortegpt.database.mongo_memory import crud
+
+
+class ChunkType(StrEnum):
+    START = "start"
+    TEXT = "text"
+    REFERENCE = "reference"
+    IMAGE = "image"
+    TOOL = "tool"
+    END = "end"
+    ERROR = "error"
+
+
+ContentType: TypeAlias = str | int
+
+
+class ResponseChunk(BaseModel):
+    type: ChunkType
+    content: ContentType | list[ContentType] | None
+    images: list[str] | None = None  # Nuevo campo para imágenes
+
+
+async def stream(agent: Agent, prompt: str, conversation_id: UUID):
+    yield ResponseChunk(type=ChunkType.START, content="")
+
+    conversation = await crud.get_conversation(conversation_id)
+
+    if conversation is None:
+        raise ValueError("Conversation not found")
+
+    conversation.add(role="user", content=prompt)
+
+    history = conversation.to_openai_format(agent.message_limit, langchain_compat=True)
+    async for content in agent.stream(history):
+        yield ResponseChunk(type=ChunkType.TEXT, content=content)
+
+    if (tool_id := ctx.tool_id.get()) is not None:
+        tool_buffer = ctx.tool_buffer.get()
+        assert tool_buffer is not None
+
+        tool_name = ctx.tool_name.get()
+        assert tool_name is not None
+
+        yield ResponseChunk(type=ChunkType.TOOL, content=None)
+
+        buffer_dict = json.loads(tool_buffer)
+
+        result, images = await agent.tool_map[tool_name](**buffer_dict)
+        
+        # Enviar imágenes si existen
+        if images:
+            yield ResponseChunk(type=ChunkType.IMAGE, content=images)
+
+        conversation.add(
+            role="assistant",
+            tool_calls=[
+                {
+                    "id": tool_id,
+                    "type": "function",
+                    "function": {
+                        "name": tool_name,
+                        "arguments": tool_buffer,
+                    },
+                }
+            ],
+        )
+        conversation.add(role="tool", content=result, tool_call_id=tool_id)
+
+        history = conversation.to_openai_format(agent.message_limit, langchain_compat=True)
+        async for content in agent.stream(history, {"tools": None}):
+            yield ResponseChunk(type=ChunkType.TEXT, content=content)
+
+    conversation.add(role="assistant", content=ctx.buffer.get())
+
+    await conversation.replace()
+
+    yield ResponseChunk(type=ChunkType.END, content="")