How does it retrieve crate documentation?
It runs `cargo docs` locally, then parses the generated HTML documentation into markdown. This information is cached in the `.docs-cache` folder in the project root.
@terhechte
基本信息
A MCP server to allow the LLM in Cursor to access Rust Analyzer, Crate Docs and Cargo Commands.
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 Rust Tools is an MCP server that enables AI agents in Cursor to access Rust type information from the Rust Analyzer LSP, retrieve crate documentation, and run Cargo commands. It includes a UI for configuration and is aimed at Rust developers using Cursor’s AI features.
Install via cargo install --git https://github.com/terhechte/cursor-rust-tools. Run with UI using cursor-rust-tools or headlessly with cursor-rust-tools --no-ui. Configure projects in ~/.cursor-rust-tools and add the generated mcp.json to your project’s .cursor/ directory to connect it with Cursor.
tokio::spawn)cargo test and cargo check from the AIcargo check from chatIt runs `cargo docs` locally, then parses the generated HTML documentation into markdown. This information is cached in the `.docs-cache` folder in the project root.
Cursor Rust Tools spins up its own Rust Analyzer instance, separate from the one in the editor. This avoids conflicts because Rust Analyzer is designed for a single consumer (e.g., open/close document ordering).
Yes. In the `~/.cursor-rust-tools` configuration, you can add an `ignore_crates` list with crate names you don’t want indexed (e.g., if they are too large).
The tool generates a `mcp.json` file (via the UI or printed to terminal). Save it in your project’s `.cursor/` directory. When Cursor detects the new file, it will prompt you to enable the MCP server. You can verify it under Cursor’s MCP settings.
Open todos include: creating a Zed extension, proper shutdown without errors, improved project removal in the UI, exposing more LSP commands, and allowing Grit operations. Only features explicitly listed in the README are supported.
OrangePro local-first CLI + MCP server for behavior mapping, grounded test generation, and dynamic proof.

Nebelus is an enterprise agentic-AI platform for regulated industries. This MCP server exposes the Nebelus Construction API: ~48 tools that give an MCP client full agent build parity — create/update/probe agents, edit ag
Gives AI agents public URLs (tunnels) for localhost, live HTTP traffic inspection, snapshot publishing, and access control. Part of LocalCan, the ngrok alternative for Mac, Windows and Linux. Free plan.

Local-first credential broker for MCP. AI coding agents (Claude Code, Cursor, Codex…) call APIs, SSH and databases through scoped presets; secrets stay in your OS keychain and never enter the agent's context.

Automated codebase documentation for GitHub: searchable docs, MCP context for AI agents, doc drift detection, and Friday Cleanup PRs.
评论