Sponsored by Deepsite.site

Cursor Rust Tools

Created By
terhechte8 months ago
A MCP server to allow the LLM in Cursor to access Rust Analyzer, Crate Docs and Cargo Commands.
Content

Cursor Rust Tools

A MCP server to allow the LLM in Cursor to access Rust Analyzer, Crate Docs and Cargo Commands.

Includes an UI for configuration.

media/example.png

What it does

Currently, various AI agents don't offer the AI the ability to access Rust type information from the LSP. This is a hurdle because instead of seeing the type, the LLM has to reason about the potential type.

In addition, the only information about the dependencies (say tokio) are what they were trained on which is out of date and potentially for a different version. This can lead to all kinds of issues.

Cursor Rust Tools makes these available over the Model Context Protocol (MCP).

  • Get the documentation for a crate or for a specific symbol in the crate (e.g. tokio or tokio::spawn)
  • Get the hover information (type, description) for a specific symbol in a file
  • Get a list of all the references for a specific symbol in a file
  • Get the implementation of a symbol in a file (retrieves the whole file that contains the implementation)
  • Find a type just by name in a file the project and return the hover information
  • Get the output of cargo test
  • Get the output of cargo check

media/screenshot.png

How it works

For the LSP functionality src/lsp it spins up a new Rust Analyzer that indexes your codebase just like the on running in your editor. We can't query the one running in the editor because Rust Analyzer is bound to be used by a single consumer (e.g. the open document action requires a close document in the right order, etc)

For documentation, it will run cargo docs and then parse the html documentation into markdown locally. This information is stored in the project root in the .docs-cache folder.

Installation

cargo install --git https://github.com/terhechte/cursor-rust-tools

Run With UI

cursor-rust-tools

This will bring up a UI in which you can add projects, install the mcp.json and see the activity.

Run Without UI

Alternatively, once you have a ~/.cursor-rust-tools set up with projects, you can also just run it via

cursor-rust-tools --no-ui

Configuration

In stead of using the UI to create a configuration, you can also set up ~/.cursor-rust-tools yourself:

[[projects]]
root = "/Users/terhechte/Developer/Rust/example1"
ignore_crates = []

[[projects]]
root = "/Users/terhechte/Developer/Rust/example2"
ignore_crates = []

ignore_crates is a list of crate dependency names that you don't want to be indexed for documentation. For example because they're too big.

Configuring Cursor

One the app is running, you can configure Cursor to use it. This requires multiple steps.

  1. Add a project-dir/.cursor/mcp.json to your project. The Cursor Rust Tools UI has a button to do that for you. Running it without UI will also show you the mcp.json contents in the terminal.
  2. As soon as you save that file, Cursor will detect that a new MCP server has been added and ask you to enable it. (in a dialog in the bottom right).
  3. You can check the Cursor settings (under MCP) to see where it is working correctly
  4. To test, make sure you have Agent Mode selected in the current Chat. And then you can ask it to utilize one of the new tools, for example the cargo_check tool.
  5. You might want to add cursor rules to tell the LLM to prefer using these tools whenever possible. I'm still experimenting with this.

media/cursor.png

The contents of all the mcp.json is the same. Cursor Rust Tools figures out the correct project via the filepath

Open Todos

  • Create a Zed extension to allow using this
  • Proper shutdown without errors
  • Removing a project is a bit frail right now (in the UI)
  • Expose more LSP commands
  • Allow the LLM to perform Grit operations
Recommend Servers
TraeBuild with Free GPT-4.1 & Claude 3.7. Fully MCP-Ready.
Playwright McpPlaywright MCP server
Visual Studio Code - Open Source ("Code - OSS")Visual Studio Code
AiimagemultistyleA Model Context Protocol (MCP) server for image generation and manipulation using fal.ai's Stable Diffusion model.
CursorThe AI Code Editor
Zhipu Web SearchZhipu Web Search MCP Server is a search engine specifically designed for large models. It integrates four search engines, allowing users to flexibly compare and switch between them. Building upon the web crawling and ranking capabilities of traditional search engines, it enhances intent recognition capabilities, returning results more suitable for large model processing (such as webpage titles, URLs, summaries, site names, site icons, etc.). This helps AI applications achieve "dynamic knowledge acquisition" and "precise scenario adaptation" capabilities.
Howtocook Mcp基于Anduin2017 / HowToCook (程序员在家做饭指南)的mcp server,帮你推荐菜谱、规划膳食,解决“今天吃什么“的世纪难题; Based on Anduin2017/HowToCook (Programmer's Guide to Cooking at Home), MCP Server helps you recommend recipes, plan meals, and solve the century old problem of "what to eat today"
ChatWiseThe second fastest AI chatbot™
Tavily Mcp
WindsurfThe new purpose-built IDE to harness magic
MCP AdvisorMCP Advisor & Installation - Use the right MCP server for your needs
Serper MCP ServerA Serper MCP Server
TimeA Model Context Protocol server that provides time and timezone conversion capabilities. This server enables LLMs to get current time information and perform timezone conversions using IANA timezone names, with automatic system timezone detection.
Jina AI MCP ToolsA Model Context Protocol (MCP) server that integrates with Jina AI Search Foundation APIs.
BlenderBlenderMCP connects Blender to Claude AI through the Model Context Protocol (MCP), allowing Claude to directly interact with and control Blender. This integration enables prompt assisted 3D modeling, scene creation, and manipulation.
Amap Maps高德地图官方 MCP Server
Context7Context7 MCP Server -- Up-to-date code documentation for LLMs and AI code editors
EdgeOne Pages MCPAn MCP service designed for deploying HTML content to EdgeOne Pages and obtaining an accessible public URL.
DeepChatYour AI Partner on Desktop
Baidu Map百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
MiniMax MCPOfficial MiniMax Model Context Protocol (MCP) server that enables interaction with powerful Text to Speech, image generation and video generation APIs.