- Home
- MCP servers
- Tree-Sitter
Tree-Sitter
- typescript
18
GitHub Stars
typescript
Language
5 months ago
First Indexed
2 months ago
Catalog Refreshed
Documentation & install
Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.
Installation
Add the following to your MCP client configuration file.
Configuration
View docs{
"mcpServers": {
"nendotools-tree-sitter-mcp": {
"command": "npx",
"args": [
"@nendo/tree-sitter-mcp",
"--mcp"
]
}
}
}Tree-Sitter MCP provides fast semantic code search and analysis for many languages, and it can run as a local MCP server for AI tools. You can search for functions and classes, trace code usage, and assess code quality across your project with sub-100ms results by parsing structure rather than scanning text.
How to use
You run the MCP server locally and connect your MCP client (for example Claude or other AI assistants) to it. Use the server to search for code elements, find all usages, and run quality and structure analyses. The MCP server exposes practical tools for code comprehension and refactoring, and it can be started from the command line and then used by your editor or AI tools without needing to modify your project setup.
How to install
Prerequisites: Node.js and npm should be installed on your system. You will install the MCP package globally so you can run its commands from anywhere.
Install the MCP package globally with npm.
npm install -g @nendo/tree-sitter-mcp
Configuration for Claude Desktop integration
Configure your Claude Desktop client to use the MCP server by adding an MCP server entry that runs the MCP tool locally.
If you prefer a manual configuration, add the following MCP server entry to your Claude Desktop config.
{
"mcpServers": {
"tree-sitter-mcp": {
"command": "npx",
"args": ["@nendo/tree-sitter-mcp", "--mcp"],
"cwd": "/path/to/your/project"
}
}
}
Starting and using the MCP server
Start the MCP server as part of your editor workflow or via your MCP client using the configured command. The server runs locally and listens for requests from the client, enabling fast semantic searches, usage tracing, and quality analyses across your codebase.
Notes on usage patterns
- Use semantic search to locate functions, classes, and variables by name across multiple languages. - Trace where a function or class is used before making changes. - Run quality analyses to identify complex functions, dead code, and architectural issues. - The server is designed to be fast by parsing code structure instead of scanning raw text.
Configuration details
The MCP server configuration used for local Claude Desktop integration is exposed here for quick setup. If you need to adjust the working directory, modify the cwd value to point to your project folder.
Available tools
search_code
Search for functions, classes, or variables by name across many languages and return matching definitions and locations.
find_usage
Identify all usages of a given identifier across the project to understand impact and refactoring scope.
analyze_code
Perform code quality and structure analyses to reveal potential architectural issues, dead code, and complex functions.
check_errors
Detect syntax errors with detailed context to help you fix issues quickly.