Code Scanner Server
An MCP server that scans codebases to extract structural information (classes, functions, etc.) with flexible filtering options and outputs in LLM-friendly formats.
README
code-scanner-server
A CLI tool and MCP server that scans code files for definitions (classes, functions, etc.), respects .gitignore, provides line numbers, and outputs LLM-friendly formats (XML/Markdown).
This project provides a versatile code scanning tool built with TypeScript and Node.js. It leverages the Tree-sitter parsing library to analyze source code and extract structural information. It can operate both as a command-line interface (CLI) tool and as an MCP (Model Context Protocol) server.
Note: This tool is under active development. While core functionality is operational, some features or specific language parsers may not be fully tested and might contain bugs or limitations.
Features
- Code Definition Extraction: Identifies functions, classes, variables, interfaces, methods, etc.
- Multi-Language Support: Parses JavaScript (
.js,.jsx), TypeScript (.ts,.tsx), C# (.cs), PHP (.php), CSS (.css), and Python (.py) via Tree-sitter. - .gitignore Aware: Automatically respects rules defined in
.gitignorefiles. - Flexible Filtering: Filter results by definition type, modifiers (
public,private), name patterns (regex), and file path patterns. - Multiple Output Formats: Generates results in Markdown (default), XML, or JSON.
- Configurable Detail Levels: Output verbosity:
minimal,standard(default),detailed. - Dual Mode Operation: Run as a standalone CLI tool or as an integrated MCP server.
Usage Modes
1. Command-Line Interface (CLI)
Run the scanner directly from your terminal. This mode requires the --directory argument specifying the target codebase.
Basic Usage:
node build/index.js --directory /path/to/your/codebase
Common Options:
-d, --directory <path>: (Required) Absolute or relative path to the directory to scan.-p, --patterns <patterns...>: Glob patterns for file extensions (e.g.,"**/*.ts""**/*.js"). Defaults to JS, TSX, CS, PHP, CSS, PY files.-f, --format <format>: Output format (xml,markdown,json). Default:markdown.-l, --detail <level>: Level of detail (minimal,standard,detailed). Default:standard.--include-types <types...>: Only include specific definition types (e.g.,class,method).--exclude-types <types...>: Exclude specific definition types.--include-modifiers <modifiers...>: Only include definitions with specific modifiers (e.g.,public).--exclude-modifiers <modifiers...>: Exclude definitions with specific modifiers.--name-pattern <regex>: Include definitions matching a JavaScript regex pattern.--exclude-name-pattern <regex>: Exclude definitions matching a JavaScript regex pattern.--include-paths <paths...>: Additional file path patterns (glob) to include.--exclude-paths <paths...>: File path patterns (glob) to exclude.-h, --help: Display detailed help information for all options.
Example (Scan TypeScript files in src, output detailed JSON):
node build/index.js -d ./src -p "**/*.ts" -f json -l detailed
2. MCP Server Mode (scan_code tool)
If run without the --directory argument, the tool starts as an MCP server, listening for requests via standard input/output. This allows integration with MCP clients like AI assistants.
- Tool Name:
scan_code - Description: Scans a specified directory for code files and returns a list of definitions according to the provided filters.
- Input Schema: Accepts arguments corresponding to the CLI options. The
directoryproperty is required.{ "type": "object", "properties": { "directory": { "type": "string", "description": "Absolute path to the directory to scan." }, "filePatterns": { "type": "array", "items": { "type": "string" }, "description": "Glob patterns for files.", "default": ["**/*.js", ..., "**/*.py"] }, "outputFormat": { "type": "string", "enum": ["xml", "markdown", "json"], "default": "markdown" }, "detailLevel": { "type": "string", "enum": ["minimal", "standard", "detailed"], "default": "standard" }, "includeTypes": { "type": "array", "items": { "type": "string" } }, "excludeTypes": { "type": "array", "items": { "type": "string" } }, "includeModifiers": { "type": "array", "items": { "type": "string" } }, "excludeModifiers": { "type": "array", "items": { "type": "string" } }, "namePattern": { "type": "string", "description": "Regex pattern for names." }, "excludeNamePattern": { "type": "string", "description": "Regex pattern to exclude names." }, "includePaths": { "type": "array", "items": { "type": "string" } }, "excludePaths": { "type": "array", "items": { "type": "string" } } }, "required": ["directory"] } - Example Usage with AI Assistant: "Use code-scanner-server scan_code on directory /path/to/project outputting xml format."
Installation
- Prerequisites: Ensure you have Node.js and npm installed.
- Clone (Optional): If you don't have the code, clone the repository.
# git clone <repository_url> # cd code-scanner-server - Install Dependencies:
npm install - Build: Compile the TypeScript code.
This creates the executable JavaScript file atnpm run buildbuild/index.js.
Configuration (MCP Server)
To use the MCP server mode, add it to your MCP client's configuration file (e.g., claude_desktop_config.json for the desktop app or cline_mcp_settings.json for the VS Code extension).
Important: Replace /path/to/code-scanner-server in the example below with the absolute path to this project's directory on your system.
Example (claude_desktop_config.json / cline_mcp_settings.json):
{
"mcpServers": {
"code-scanner-server": {
"command": "node",
"args": [
"/absolute/path/to/your/code-scanner-server/build/index.js" // <-- Replace this path! (e.g., "C:\\Users\\YourUser\\Projects\\code-scanner-server\\build\\index.js" on Windows)
],
"env": {},
"disabled": false,
"autoApprove": [] // Add tool names here for auto-approval if desired
}
}
}
Remember to restart your MCP client application (IDE, Desktop App) after modifying the configuration for changes to take effect.
Development
- Watch Mode: Automatically rebuild the project when source files change:
npm run watch - Debugging (MCP Mode): Debugging MCP servers over stdio can be complex. Use the MCP Inspector tool for easier debugging:
This starts the server with the Node.js inspector attached and provides a URL to connect debugging tools (like Chrome DevTools).npm run inspector
Acknowledgments
This project was significantly developed with the assistance of AI, primarily using Google's Gemini 2.5 Pro model accessed via the Roo Code extension for Visual Studio Code.
License
This project is licensed under the GNU General Public License v3.0 - see the LICENSE file for details.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.