LSP MCP Server Integration

What does “LSP” MCP Server do?

The LSP MCP (Model Context Protocol) Server acts as a bridge between Language Server Protocol (LSP) servers and AI assistants. By connecting to an LSP server, it enables LLMs and other AI clients to interact with codebases through standardized LSP features. This allows AI tools to perform advanced code analysis, fetch hover documentation, receive code completion suggestions, access diagnostics, and even apply code actions directly inside an editor or development workflow. The LSP MCP Server enhances developer productivity by facilitating seamless AI-driven interactions with source code, making tasks like code navigation, error resolution, and intelligent code completion more accessible and automated.

List of Prompts

No explicit prompt templates are listed in the repository or documentation.

List of Resources

• lsp-diagnostics://
  Access real-time diagnostic messages (errors, warnings, etc.) from open files, supporting live updates via subscriptions.
• lsp-hover://
  Retrieve hover information at specific file locations, enabling context-aware code explanations.
• lsp-completions://
  Get code completion suggestions for a given position within a file, supporting enhanced coding assistance.

List of Tools

• get_info_on_location
  Fetch hover (documentation) information at a specific location in a file.
• get_completions
  Provide code completion suggestions at a specified position in a file.
• get_code_actions
  Retrieve available code actions (e.g., quick fixes) for a given range in a file.
• open_document
  Open a file in the LSP server for analysis and interaction.
• close_document
  Close an open file in the LSP server context.
• get_diagnostics
  Get diagnostic messages (errors and warnings) for currently open files.
• start_lsp
  Start the underlying LSP server with a specified root directory.
• restart_lsp_server
  Restart the LSP server instance without restarting the MCP server.
• set_log_level
  Dynamically change the logging verbosity of the server at runtime.

Use Cases of this MCP Server

• Codebase Exploration
  Developers and AI agents can query hover information and code completions, making it easier to understand and navigate large codebases.
• Automated Code Review & Error Detection
  By accessing diagnostics and code actions, AI tools can highlight errors and suggest fixes during code review or continuous integration.
• Intelligent Code Completion
  LLMs can leverage real LSP completions to provide accurate, context-aware suggestions while writing or refactoring code.
• Interactive Learning & Documentation
  AI assistants can fetch hover details and documentation on-demand, enhancing learning and onboarding for new developers.
• Editor and IDE Integration
  The server can be integrated into editors to provide real-time AI-powered features, such as suggesting code fixes or explaining code sections.

How to set it up

Windsurf

1. Ensure you have Node.js (v16+) and npm installed.
2. Locate your Windsurf configuration file.
3. Add the LSP MCP Server in the mcpServers section:

   {
     "mcpServers": {
       "lsp-mcp": {
         "type": "stdio",
         "command": "npx",
         "args": [
           "tritlo/lsp-mcp",
           "<language-id>",
           "<path-to-lsp>",
           "<lsp-args>"
         ]
       }
     }
   }
   

4. Save the configuration and restart Windsurf.
5. Verify integration by running a test LSP command.

Claude

1. Install Node.js (v16+) and npm.
2. Find the Claude configuration file.
3. Insert the LSP MCP Server in the mcpServers section:

   {
     "mcpServers": {
       "lsp-mcp": {
         "type": "stdio",
         "command": "npx",
         "args": [
           "tritlo/lsp-mcp",
           "<language-id>",
           "<path-to-lsp>",
           "<lsp-args>"
         ]
       }
     }
   }
   

4. Save changes, restart Claude, and verify by running claude --mcp-debug for logs.
5. Ensure the server starts successfully and processes LSP commands.

Cursor

1. Confirm Node.js (v16+) and npm are installed.
2. Open Cursor’s configuration file.
3. Add the LSP MCP Server:

   {
     "mcpServers": {
       "lsp-mcp": {
         "type": "stdio",
         "command": "npx",
         "args": [
           "tritlo/lsp-mcp",
           "<language-id>",
           "<path-to-lsp>",
           "<lsp-args>"
         ]
       }
     }
   }
   

4. Save and restart Cursor.
5. Test by opening a project and checking for LSP responses.

Cline

1. Install Node.js (v16+) and npm.
2. Edit the Cline configuration file.
3. Add the following:

   {
     "mcpServers": {
       "lsp-mcp": {
         "type": "stdio",
         "command": "npx",
         "args": [
           "tritlo/lsp-mcp",
           "<language-id>",
           "<path-to-lsp>",
           "<lsp-args>"
         ]
       }
     }
   }
   

4. Save and restart Cline.
5. Validate setup by running LSP-related commands in the interface.

Securing API Keys

If the LSP server or MCP setup requires API keys, use environment variables for security:

{
  "mcpServers": {
    "lsp-mcp": {
      "type": "stdio",
      "command": "npx",
      "args": [
        "tritlo/lsp-mcp",
        "<language-id>",
        "<path-to-lsp>",
        "<lsp-args>"
      ],
      "env": {
        "API_KEY": "${LSP_API_KEY}"
      },
      "inputs": {
        "api_key": "${LSP_API_KEY}"
      }
    }
  }
}

How to use this MCP inside flows

Using MCP in FlowHunt

To integrate MCP servers into your FlowHunt workflow, start by adding the MCP component to your flow and connecting it to your AI agent:

Click on the MCP component to open the configuration panel. In the system MCP configuration section, insert your MCP server details using this JSON format:

{
  "lsp-mcp": {
    "transport": "streamable_http",
    "url": "https://yourmcpserver.example/pathtothemcp/url"
  }
}

Once configured, the AI agent is now able to use this MCP as a tool with access to all its functions and capabilities. Remember to change “lsp-mcp” to whatever the actual name of your MCP server is and replace the URL with your own MCP server URL.

Overview

Our opinion

The LSP MCP Server provides comprehensive documentation, a range of tools and resources for LSP interaction, and good setup instructions. However, it lacks explicit support for prompt templates and makes no mention of sampling or roots support. Overall, it’s solid for codebase and LSP integration but might benefit from more advanced MCP features.

Rating: 7/10

MCP Score