{"$schema":"https://wellknown.network/schemas/agent-record-v1.json","schemaVersion":"1","id":"ag_sq5mknk8gf54","handle":"chat-mcp","url":"https://wellknown.network/agents/chat-mcp","links":{"self":"https://wellknown.network/agents/chat-mcp/record.json","html":"https://wellknown.network/agents/chat-mcp","markdown":"https://wellknown.network/agents/chat-mcp/record.md","api":"https://wellknown.network/api/v1/agents/chat-mcp","status":"https://wellknown.network/api/v1/agents/chat-mcp/status","claim":"https://wellknown.network/agents/chat-mcp/claim","claimApi":"https://wellknown.network/api/v1/claims","claimDescriptor":"https://wellknown.network/agents/chat-mcp/claim.json","badge":"https://wellknown.network/agents/chat-mcp/badge.svg","openapi":"https://wellknown.network/openapi.json","history":"https://wellknown.network/api/v1/agents/chat-mcp/history","tools":"https://wellknown.network/api/v1/agents/chat-mcp/tools"},"ard":{"identifier":"urn:air::server:chat-mcp","type":"application/mcp-server-card+json"},"kind":"mcp_server","declared":{"name":"chat-mcp","summary":"MCP客户端库，用于在对话中调用MCP工具","description":"# Chat-MCP\n\n**一个用于在对话中调用MCP（Model Context Protocol）工具的Python客户端库**\n\nChat-MCP 提供了一个简洁而强大的接口，让您能够轻松地将MCP服务器集成到AI agent项目中。通过两个简单的api，并配置好.env文件，就可以在开发agent和chat助手时高效的调用mcp服务。作为agentic项目的底层脚手架，分享大家使用。\n\n## 核心特性\n\n- **简洁的API设计**: 二个主要接口覆盖所有使用场景，提高agent应用开发效率\n- **优化提示词**: 提高tools调用的成功率，持续优化中\n- **异步架构**: 完全基于asyncio的高性能处理\n- **智能工具管理**: 自动工具收集、过滤和并行执行\n- **多模型支持**: 通过LiteLLM支持多种LLM提供商，支持本地模型配置\n\n## 关键技术栈\n\n- **[MCP Python SDK](https://github.com/modelcontextprotocol/python-sdk)**: Model Context Protocol的官方Python实现\n- **[LiteLLM](https://docs.litellm.ai/)**: 统一的LLM API接口库，支持100+模型\n- **[Pydantic](https://docs.pydantic.dev/)**: 数据验证和设置管理框架\n\n## 项目结构\n\n```\nchat-mcp/\n├── docs/                 # 详细的开发文档和规范\n├── src/chat_mcp/         # 主要源代码包\n│   ├── tests/            # 测试套件\n│   ├── __init__.py       # 公共API导出\n│   ├── easy_chat.py      # 简化API接口\n│   ├── ai_provider.py    # LLM提供商集成\n│   ├── mcp_service.py    # MCP服务器管理\n│   ├── mcp_chat_handler.py # 聊天处理逻辑\n│   ├── ipc_handler.py    # 进程间通信处理\n│   └── mcp_types.py      # 数据类型定义\n├── temp_arxiv_storage/   # ArXiv演示数据存储\n├── pyproject.toml        # uv项目配置\n├── uv.lock              # 依赖锁定文件\n└── main.py              # 演示入口\n```\n\n## 核心API\n\n### MCPChatTool - 主要接口\n\n```python\nfrom chat_mcp import MCPChatTool\n\n# 创建聊天工具实例\nchat_tool = MCPChatTool()\n\n# 启动MCP服务器，这里使用arxiv-mcp-server作为演示样例，可以添加多个不同的mcp server，只需要多次调用\nserver = await chat_tool.start_mcp_server(\n    server_id=\"arxiv\",\n    name=\"ArXiv Research\",\n    command=\"uv\",\n    args=[\"tool\", \"run\", \"arxiv-mcp-server\"]\n)\n\n# 进行对话\nresult = await chat_tool.chat_with_mcp(\n    user_message=\"搜索关于机器学习的最新论文\",\n    enabled_server_ids=[\"arxiv\"] # 这里可以不设置enabled_server_ids,默认所有server都会被添加\n)\n```\n\n## 安装\n\n使用 [uv](https://github.com/astral-sh/uv) (推荐):\n\n```bash\nuv add chat-mcp\n```\n\n或使用 pip:\n\n```bash\npip install chat-mcp\n```\n\n## 快速开始\n\n1. **配置环境变量** (创建 `.env` 文件):\n\nLLM配置：使用litellm规范，可以自行配置，支持大部分模型包括自定义大模型\n\n```env\nOPENAI_API_KEY=your_openai_api_key\nOPENAI_API_BASE=https://api.openai.com/v1\nLLM_MODEL=gpt-3.5-turbo\n```\n如果是vllm本…","publisher":null,"homepage":null,"repository":null,"version":"0.1.1","license":null,"protocols":["mcp"],"tags":["mcp"],"pricing":null,"endpoints":[{"url":"pypi:chat-mcp","type":"package_pypi","auth":null,"probeable":false}],"skills":null,"tools":null,"extra":null,"attribution":{"kind":"pypi","name":"pypi","summary":"pypi","version":"pypi","description":"pypi"}},"derived":{"capabilities":[{"slug":"communication.chat","name":"Team Chat","confidence":1,"provenance":"derived"}],"categories":["communication"],"language":"en"},"observed":{"status":"unknown","statusReason":"Distributed as a package to run locally; no network endpoint to check.","lastOkAt":null,"lastProbedAt":null,"statusComputedAt":null,"reliability30d":null,"latestObservations":[],"tools":null,"package":{"name":"chat-mcp","registry":"pypi","observedAt":"2026-11-02T23:27:08.184Z","publishedAt":"2025-06-01T18:26:10.447912Z","latestVersion":"0.1.1"},"toolSurface":null,"endpointFacts":[]},"verification":{"claimed":false,"claimedAt":null,"proofs":[]},"provenance":{"sources":[{"source":"pypi","key":"chat-mcp","url":"https://pypi.org/project/chat-mcp/","firstSeenAt":"2026-09-09T11:30:11.602Z","fetchedAt":"2026-11-02T23:24:17.926Z","normalizedAt":"2026-11-02T23:24:17.926Z"}]},"firstSeenAt":"2026-09-09T11:30:11.602Z","updatedAt":"2026-11-02T23:27:08.184Z"}