🦀 Rust MCP Server

Supercharge your AI assistant with professional-grade Rust intelligence.

Standard AI tools only see your files as text. Rust MCP acts as a semantic bridge, giving LLMs (like Claude, GPT-4, or local models) the same deep understanding that rust-analyzer provides to your IDE. It enables AI tools to work with Rust code idiomatically through Language Server Protocol capabilities, avoiding brittle string manipulation and providing intelligent code analysis.

Warning

This project is currently in Alpha. It is functional and useful, but expect breaking changes and occasional rough edges. However, most core tools are fully tested and ready for production-level development tasks. We welcome feedback and contributions!

🌟 Why this server?

While other MCP servers provide basic file access, Rust MCP focuses on Semantic Context:

• Context-Aware Refactoring: Rename symbols reliably using code snippets instead of brittle line numbers.
• Smart Navigation: Find real definitions and references, not just text matches.
• Compiler-Grade Feedback: Get actual cargo check diagnostics directly in your chat.
• Efficiency: Outline large files with document_symbols to save LLM tokens.

Quick Start

1. Build: cargo build --release
2. Configure: See the Configuration section below.
3. Use: Through AI assistants with natural language prompts like "Show me the definition of the main function".

Features - Core Tool Suite

Code Analysis & Navigation (Context-Aware)

• get_hover - Get symbol signature and documentation.
• get_symbol_source - Get source code of specific symbol.
• document_symbols - Get file structure (outline) - Recommended for large files.
• find_definition - Navigate to symbol definitions.
• find_references - Find all symbol uses.
• get_diagnostics - Get compiler errors/warnings for a specific file.
• workspace_symbols - Search project symbols.
• get_type_hierarchy - Get type relationships for symbols.

Refactoring

• rename_symbol - Rename with scope awareness (context-aware).
• extract_function - (Experimental) Extract code into functions.
• inline_function - (Experimental) Inline function calls.

Quality Assurance & Project Management

• run_cargo_check - Execute cargo check with full error parsing.
• apply_clippy_suggestions - (In Progress) Apply clippy automatic fixes.

Prerequisites

• Rust toolchain (1.70+)
• rust-analyzer installed (defaults to ~/.cargo/bin/rust-analyzer)
• An MCP-compatible client (Claude Desktop, Gemini CLI, Roo-Code, etc.)

Installation

git clone https://github.com/dexwritescode/rust-mcp
cd rust-mcp
cargo build --release

The server binary will be available at target/release/rustmcp.

Configuration

Detailed guides for setting up Rust MCP with various clients:

• Claude Desktop
• Gemini CLI
• Roo-Code (VS Code)
• ESP32 & Custom Toolchains
• Other MCP Clients (Cursor, etc.)
• Environment Variables

💡 Usage Examples

These examples show how you can interact with your AI assistant once Rust MCP is configured.

🔍 Deep Code Analysis

• "Find the definition of the AppState struct and show me its source code."
• "What does the handle_request function do? Show me its signature and documentation."
• "Where else is the UserRegistry trait used or implemented in this workspace?"
• "Give me a high-level outline of src/lib.rs so I can understand its structure."

🛠 Semantic Refactoring

• "I want to rename the internal field count to total_processed. Here is the code block where it's defined..."
• "Please rename the Storage trait to DataStore across the entire project."
• "Help me rename this local variable idx to index to improve readability."

⚡ Troubleshooting & Project Health

• "Run a full cargo check and tell me if my recent changes introduced any new warnings."
• "Check the current file for any borrow checker errors."
• "Show me the type hierarchy for MyCustomError to see which traits it implements."

Architecture

The server is built with a modular architecture:

• src/analyzer/ - rust-analyzer LSP client integration.
• src/server/ - MCP server implementation and tool handlers.
• src/tools/ - Modular tool logic.

Compiler Safety Limits and Errors

To keep the server responsive, compilation helpers run with guardrails (30s timeout, 1MB output cap). See Compiler Safety for more info.

Troubleshooting

See the Troubleshooting section in individual guides or check that rust-analyzer is installed and accessible.

Contributing

1. Fork the repository
2. Create a feature branch
3. Implement your changes with tests
4. Submit a pull request

---

Built with ❤️ for the Rust & AI ecosystem.