Shy2593666979/AgentChat

What it solves

AgentChat is a modern intelligent dialogue system designed to move beyond simple chatbots by providing a comprehensive framework for AI agents that can reason, make decisions, and interact with external tools and knowledge bases. It addresses the need for a unified platform where multiple AI agents can collaborate to automate complex tasks.

How it works

The system uses a decoupled frontend (Vue 3) and backend (FastAPI) architecture. It leverages LangChain to orchestrate LLMs and implements a three-layer memory architecture (short-term, historical summary, and long-term) to maintain context and user preferences. It integrates RAG (Retrieval-Augmented Generation) for knowledge base queries and supports the Model Context Protocol (MCP) for dynamic tool loading.

Who it’s for

It is intended for developers and organizations looking to deploy a full-stack AI agent platform with built-in support for tool calling, knowledge retrieval, and multi-agent collaboration.

Highlights

• Multi-Agent Collaboration: Supports sub-agents that can work together to achieve goals.
• HITL MCP Generation: Uses a Human-In-The-Loop mechanism to conversationally generate MCP servers from OpenAPI specifications.
• Three-Layer Memory: Intelligent management of context using short-term memory, automated summaries, and persistent long-term preferences.
• RAG Integration: Built-in knowledge base system with semantic chunking and vector retrieval.
• Tool Ecosystem: Supports multi-turn tool calling (sequential dependency) and custom tool extensions via Swagger/OpenAPI.
• Task Visualization: Provides real-time task flowcharts to visualize the agent's planning process.