🎯 Overview
AgentChat is a 是一个基于 LLM 的智能体交流平台，内置默认 Agent 并支持用户自定义 Agent。通过多轮对话和任务协作，Agent 可以理解并协助完成复杂任务。项目集成 LangChain、Function Call、MCP 协议、RAG、Memory、HITL、Skill、Milvus 和 ElasticSearch 等技术，实现高效的知识检索与工具调用，使用 FastAPI 构建高性能后端服务。
⚙️ Core Capabilities
- AGENT: Leverages agent paradigms for superior performance.
- AGENT-MEMORY: Designed to work seamlessly within modern agent-memory ecosystems.
- AGENT-SKILLS: Provides dedicated abstractions for agent-skills architectures.
- HUMAN-IN-THE-LOOP: Optimized for human-in-the-loop-based execution pipelines.
- Production Ready: Extensively tested to prevent edge-case failures.
🛠️ Developer Integration
As a lightweight solution, AgentChat provides high-level APIs for fast deployment. It effectively reduces technical debt by enforcing standardized workflows.
📈 Production Benefits
Unlike generic alternatives, it offers a laser-focused feature set that solves exact domain problems. The vibrant open-source ecosystem around it guarantees long-term support.