leptos
MCP server from kneiht/leptos-mcp-server
claude mcp add --transport stdio kneiht-leptos-mcp-server ./target/release/leptos-mcp-server
How to use
This MCP server provides Leptos documentation and code-analysis tooling via a JSON-RPC over stdio interface. It exposes three tools: list-sections to enumerate all available Leptos documentation sections with use cases; get-documentation to retrieve the content of a specific documentation section by name (e.g., signals, views, routing); and leptos-autofixer to analyze Leptos code and suggest fixes for common issues. Clients send JSON-RPC requests to the server and receive responses containing the tool results or content. Typical usage involves sending a tools/list or tools/call request with the appropriate arguments, then parsing the returned data to present to the user or feed into agents.
To interact, start the server and communicate using the JSON-RPC 2.0 over stdio protocol as demonstrated in the Testing section of the README. For example, you can request the list of sections, fetch documentation for a specific section like signals, or run the autofixer to obtain suggested code improvements for a Leptos project.
How to install
Prerequisites:
- Rust toolchain (rustup, cargo) installed on your system
- Git (optional, for cloning the repository)
Install / build from source:
- Clone the repository: git clone <repository-url> leptos-mcp-server cd leptos-mcp-server
- Build in release mode for best performance: cargo build --release
- Run the server (produces the executable at target/release/leptos-mcp-server): cargo run --release # or run the binary directly from target/release/
- Verify the server starts and is ready to accept JSON-RPC over stdio as described in the README.
Notes:
- If you clone a fork or mirror, ensure the project builds with cargo build --release before attempting to run the server.
Additional notes
Tips and common issues:
- The server communicates via JSON-RPC 2.0 over stdio. Ensure your client adheres to the protocol (methods like tools/list and tools/call with proper arguments).
- The executable path in production should be the compiled binary (e.g., ./target/release/leptos-mcp-server). Adjust your mcp_config command accordingly if you place the binary in a different location.
- If the server fails to start, confirm Rust toolchain is installed and the project dependencies compiled successfully with cargo build --release.
- For get-documentation calls, the section name should match the documents listed in the Documentation Sections table (e.g., signals, views, routing). -Environment variables are currently not required by default; if you introduce config via environment later, you can extend the mcp_config with an env object per server.
- The tooling set can be extended by adding new commands; refer to the README’s features table to see available tool names.
Related MCP Servers
goose
an open source, extensible AI agent that goes beyond code suggestions - install, execute, edit, and test with any LLM
cunzhi
告别AI提前终止烦恼,助力AI更加持久
probe
AI-friendly semantic code search engine for large codebases. Combines ripgrep speed with tree-sitter AST parsing. Powers AI coding assistants with precise, context-aware code understanding.
mcp-center
A centralized platform for managing and connecting MCP servers. MCP Center provides a high-performance proxy service that enables seamless communication between MCP clients and multiple MCP servers.
backlog -rust
MCP server for Backlog, project management service.
perplexity-web-api
🔍 Perplexity AI MCP without API key