Multi-MCP AI Agent
by aviban15
The Multi-MCP AI Agent is an agentic AI system that utilizes multiple MCP servers to provide a wide range of capabilities, including external service integrations and real-time communication. It includes a Telegram bot interface and Server-Sent Events (SSE) for live updates.
Last updated: N/A
What is Multi-MCP AI Agent?
The Multi-MCP AI Agent is an agentic AI system designed to leverage multiple Model Control Protocol (MCP) servers. This architecture allows the agent to perform a diverse set of tasks, ranging from basic calculations to complex integrations with external services like Google Workspace and web scraping.
How to use Multi-MCP AI Agent?
To use the agent, first clone the repository and set up a virtual environment using UV. Install the necessary dependencies and configure environment variables for API keys and credentials. Configure MCP servers and model configurations in the respective YAML and JSON files. Start the agent and Telegram bot server using the provided UV commands.
Key features of Multi-MCP AI Agent
Multi-MCP Architecture
Cognitive Modules (Perception, Decision-Making, Memory, Action)
External Service Integration (Google Workspace, Web Scraping, DuckDuckGo)
Real-time Communication (Telegram Bot, Server-Sent Events)
Agent loop management, Session handling, Context management, Strategic decision making
Use cases of Multi-MCP AI Agent
Automated document processing and semantic search
Web content extraction and data analysis
Integration with Google Workspace for email, spreadsheets, and drive management
Real-time interaction and task execution through a Telegram bot
FAQ from Multi-MCP AI Agent
What is an MCP server?
What is an MCP server?
An MCP (Model Control Protocol) server is a component that provides specific functionalities or models to the agent. The agent can utilize multiple MCP servers to access a wider range of capabilities.
How do I configure the Google Workspace integration?
How do I configure the Google Workspace integration?
You need to generate credentials.json using Google OAuth client and place it in the root directory. Run the application once to generate token.json.
What is the role of the Telegram bot?
What is the role of the Telegram bot?
The Telegram bot provides a user interface for interacting with the agent. You can send commands and receive real-time responses through the bot.
What kind of data is streamed through SSE?
What kind of data is streamed through SSE?
The SSE (Server-Sent Events) server streams real-time updates and events from the agent, allowing clients to receive live information about the agent's activities.
How does the agent make decisions?
How does the agent make decisions?
The agent uses a cognitive architecture with modules for perception, memory, decision-making, and action. The decision module makes strategic decisions based on input, context, and available tools.