Quirox Telegram Bot Backend

What is Quirox Telegram Bot Backend?

Quirox Telegram Bot Backend provides backend services for a Telegram bot, featuring an AI agent powered by Google Gemini that can interact with tools like Google Sheets, web search (DuckDuckGo), and web content fetching. It uses the Model Context Protocol (MCP) to manage communication between the AI model and the tools, and exposes a FastAPI web server.

How to use Quirox Telegram Bot Backend?

Set up a .env file with your Google Gemini API key and Google Cloud credentials (service account or OAuth). Run python mcp_server.py in one terminal, then python main.py in another. Send POST requests to the /chat endpoint with a JSON body containing a "message" field to interact with the AI agent.

Key features of Quirox Telegram Bot Backend

• AI agent using Google Gemini Flash model for reasoning and task execution.
• Google Sheets integration: create, read, write, manage sheets, share.
• Web search via DuckDuckGo and web content fetching from URLs.
• FastAPI web server exposing a /chat endpoint for HTTP interactions.
• MCP framework managing communication between AI model and tools.

Use cases of Quirox Telegram Bot Backend

• Automate spreadsheet creation, data entry, and sheet management via chat.
• Perform web searches and retrieve content from URLs through conversational commands.
• Extend a Telegram bot with AI-powered tool interactions for document or data workflows.

FAQ from Quirox Telegram Bot Backend

What are the prerequisites?

Python 3.8+, access to Google Cloud Platform for Sheets/Drive API, and a Google Gemini API key.

How do I authenticate with Google services?

You can use a service account (recommended for servers) or OAuth 2.0. Provide the credentials via environment variables in the .env file.

How do I interact with the backend?

Send POST requests to http://localhost:8000/chat (or your server’s address) with a JSON body containing a "message" field.

Are Gmail tools available?

Gmail functions are present in the code but currently commented out and not exposed in the running server.

What dependencies does the project require?

Key dependencies include fastapi, uvicorn, mcp, google-api-python-client, google-generativeai, httpx, beautifulsoup4, and others listed in requirements.txt.