Get the FREE Ultimate OpenClaw Setup Guide →

Language -Bridge

A universal VS Code extension that bridges any Language Server Protocol (LSP) capabilities to MCP tools and GitHub Copilot Language Model Tools, enabling intelligent code navigation and analysis across all programming languages.

Installation
Run this command in your terminal to add the MCP server to Claude Code.
Run in terminal:
Command
claude mcp add --transport stdio sehejjain-language-server-mcp-bridge node path/to/server.js \
  --env VSODEV="placeholder-for-vscode-extension-env" \
  --env EXTENSION_ID="sehejjain.lsp-mcp-bridge"

How to use

Language Server tools for Copilot bridges the capabilities of any active Language Server Protocol (LSP) in VS Code with MCP tooling and GitHub Copilot. The extension exposes a set of 10 LSP-powered tools that Copilot can query automatically, such as lsp_definition for symbol definitions, lsp_references for references, lsp_hover for documentation, and lsp_completion for code completion. These tools enable Copilot to reason about code structure, navigate large codebases, and provide more accurate answers or suggestions by leveraging the underlying language servers in your editor.

To use the tools, install the LSP MCP Bridge extension in VS Code and ensure you have a working language server configured in your environment (for example clangd for C/C++, Pyright/Pylance for Python, tsserver for TypeScript/JavaScript, rust-analyzer for Rust, etc.). Once installed, Copilot will automatically discover and use the 10 tools without additional configuration. You can also test the tools manually via the Command Palette with commands like "LSP MCP: Test MCP Tools at Cursor" or "LSP MCP: List Language Model Tools" to see the available capabilities and confirm they respond to your cursor position.

In practice, you can ask Copilot questions such as “What does this function do?” or request a definition, references, or a hover description for a symbol. The bridge mechanisms translate those requests into LSP queries and return structured results, enabling Copilot to provide deeper, context-aware answers across many programming languages.

How to install

Prerequisites:

  • Visual Studio Code installed
  • Node.js (recommended LTS) installed
  • An active language server configured in VS Code for your project (e.g., clangd, Pylance, tsserver, rust-analyzer, etc.)

Installation steps:

  1. Install the extension from the VS Code Marketplace:
  1. Install from the command line (optional):
code --install-extension sehejjain.lsp-mcp-bridge
  1. If you are developing locally, clone the repository and install dependencies:
git clone <repository-url>
cd Language-Server-MCP-Bridge
npm install
  1. Build/package (for development):
npm run compile
npx @vscode/vsce package
  1. Load the packaged extension in VS Code or install via the generated .vsix file if you are packaging for distribution.

Tip: Ensure your language servers are configured and usable within VS Code before relying on the MCP bridge for Copilot interactions.

Additional notes

Notes and tips:

  • The bridge relies on VS Code's existing language servers; you don’t need to install new servers specifically for MCP unless you want additional language support.
  • If Copilot does not seem to use the tools, verify that the VS Code MCP system is enabled and the extension is properly loaded (check the Extensions view and Output logs).
  • For troubleshooting, use the Command Palette to run:
    • "LSP MCP: Test MCP Tools at Cursor" to validate tool responses at the cursor position
    • "LSP MCP: List Language Model Tools" to confirm the 10 LSP tools are registered
  • Environment variables can be used to customize behavior or pass metadata to the MCP bridge during startup.
  • If you package and distribute this as an npm-based server, ensure the entry point script path to server.js is correct in the mcp_config.

Related MCP Servers

Sponsor this space

Reach thousands of developers