example.py

"""
deepset Haystack integration example.

Prerequisites:
    cp .env.example .env
    pip install -r requirements.txt
"""

import asyncio
import os

from dotenv import load_dotenv
from haystack import Pipeline
from haystack.components.generators.chat import OpenAIChatGenerator
from haystack.components.tools import ToolInvoker
from haystack.dataclasses import ChatMessage

from metorial import Metorial, metorial_haystack

load_dotenv()


async def main():
  metorial = Metorial(api_key=os.getenv("METORIAL_API_KEY"))

  # Create a deployment for Metorial Search — built-in web search, no auth needed
  deployment = metorial.provider_deployments.create(
    name="Metorial Search",
    provider_id="metorial-search",
  )

  session = await metorial.connect(
    adapter=metorial_haystack(),
    providers=[
      {"provider_deployment_id": deployment.id},
      # (Optional) Add an OAuth provider like Slack or GitHub:
      # {"provider_deployment_id": "your-slack-deployment-id", "provider_auth_config_id": "auth-config-id"},
    ],
  )
  tools = session.tools()

  generator = OpenAIChatGenerator(model="gpt-4o", tools=tools)
  tool_invoker = ToolInvoker(tools=tools)

  pipeline = Pipeline()
  pipeline.add_component("generator", generator)
  pipeline.add_component("tool_invoker", tool_invoker)
  pipeline.connect("generator.replies", "tool_invoker.messages")

  messages = [
    ChatMessage.from_user(
      "Search the web for the latest news about AI agents and summarize the top 3 stories."
    )
  ]
  result = pipeline.run({"generator": {"messages": messages}})

  print(result["tool_invoker"]["tool_messages"])


if __name__ == "__main__":
  asyncio.run(main())