What are the prerequisites to run Cursor History MCP 📜?
You need Docker, Python 3.8 or higher, FastAPI, LanceDB, and Ollama installed.
@Nossim
基本信息
API service to search vectorized Cursor IDE chat history using LanceDB and Ollama
使用下面的配置,将此服务器添加到你的 MCP 客户端。
{
"mcpServers": {
"Cursor-history-MCP": {
"command": "docker",
"args": [
"build",
"-t",
"cursor-history-mcp",
"."
]
}
}
}工具列表暂未提供
当前收录暂时没有可展示的工具列表,不代表服务没有工具。请查阅项目文档了解支持的能力。
工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。
Cursor History MCP is an API service that enables vectorized search through chat history from the Cursor IDE. It uses LanceDB for vector storage and Ollama for local LLM processing, and is designed for self-hosted deployment.
Clone the repository, build the Docker image (docker build -t cursor-history-mcp .), and run the container (docker run -p 8000:8000 cursor-history-mcp). Access the API documentation at http://localhost:8000/docs. Use the POST /search endpoint to query chat history and GET /history to retrieve all records.
You need Docker, Python 3.8 or higher, FastAPI, LanceDB, and Ollama installed.
Clone the repository, build the Docker image, and run the container. See the Getting Started section for exact commands.
Send a POST request to the `/search` endpoint with a JSON body containing your query string.
Send a GET request to the `/history` endpoint to get the full list of chat messages.
The service is self-hosted, meaning you can run it locally or on your own server, keeping your data under your control.

Catch intent conflicts before code conflicts. The open-source coordination protocol for coding agents, built above Git.

MCP server for AI Agents — exposes Yocoolab feedback threads and design selections as tools
ZeroBeacon.ai — Collision-proof commerce router for AI agents. 1000 MCP tools: commerce, trust, billing, sovereign ID + everyday tools. LIVE.

AI SEO tools for MCP clients — audit whether AI crawlers can read a website, validate JSON-LD and robots.txt, and generate llms.txt, AI-crawler-aware robots.txt, and JSON-LD schema without leaving the conversation.

One case file per task — shared by every session, agent and person that touches it. Self-hosted MCP server and live board. MIT.
评论