Codegraphx (cgx)
@techcraze00
Codegraphx (cgx) について
CodeGraphX (CGX) — a local, token-efficient codebase graph engine & MCP server for AI coding agents. Tree-sitter parsing, a bi-temporal SQLite semantic graph, O(1) symbol lookup, and impact/blast-radius tracing so agents answer 'what breaks if I change this?' in a few hundred tok
基本情報
設定
以下の設定を使って、このサーバーを MCP 対応クライアントに追加してください。
{
"mcpServers": {
"codegraphx": {
"description": "CodeGraphX MCP server for codebase intelligence. Enables AI assistants to query indexed source code, analyze dependencies, trace symbol impact, verify implementation tasks, inspect repository changes, and navigate large codebases without expensive file scanning.",
"command": "npx",
"args": [
"-y",
"-p",
"codegraphx",
"cgx-mcp"
]
}
}
}ツール
ツールは検出されませんでした
ツールは README から自動的に抽出されます。メンテナーは ## Tools という見出しの下に記載することで、このタブに反映できます。
概要
What is CodeGraphX?
CodeGraphX (CGX) is a local, token-efficient codebase graph engine for AI coding agents and human developers. It parses source code with Tree-sitter, stores a bi-temporal semantic graph in SQLite, and serves it over a CLI and an MCP server.
How to use CodeGraphX?
Install with npm install -g codegraphx, then run cgx init to index the codebase. Query symbols with cgx query <symbol> or trace impact with cgx impact. The MCP server (cgx-mcp) exposes six tools for AI agents, with zero-config auto-indexing on first start.
Key features of CodeGraphX
- Persistent, queryable graph of code symbols and relationships
- O(1) symbol lookup via Bloom filter
- Recursive upstream/downstream impact tracing
- Cross-language call graph linking (frontend to backend routes)
- Bi-temporal history: query codebase as of any commit
- Token-optimized output in TOON format for AI agents
Use cases of CodeGraphX
- Identifying blast radius of a function change before refactoring
- Providing structural context to AI agents without scanning files
- Detecting dead imports, circular dependencies, and syntax errors
- Comparing task descriptions against actual code changes for verification
- Understanding cross-language API contracts between frontend and backend
FAQ from CodeGraphX
How does CodeGraphX differ from a Language Server or grep?
A Language Server answers "go to definition" for editors; grep finds text, not structure. CodeGraphX answers structural, whole-repo reasoning questions like blast radius and dependency cycles, optimized for AI agents with token-efficient output.
Is CodeGraphX cloud-based or local?
100% local. No cloud, no telemetry, no network — code never leaves the machine.
What languages does CodeGraphX support?
Python, JavaScript, JSX, TypeScript, TSX, HTML, and CSS. It extracts functions, classes, calls, imports, and framework-specific patterns (Flask/FastAPI, Express, fetch/axios).
How do I update the graph after code changes?
Run cgx init or cgx scan to perform a full re-index (unchanged files are served from cache). Use cgx watch for automatic re-indexing on file changes.
What are the runtime requirements?
Node.js version 18 or higher, running in a CommonJS environment.
「その他」の他のコンテンツ
FastMCP v2 🚀
jlowin🚀 The fast, Pythonic way to build MCP servers and clients.
Blender
ahujasidOpen-source MCP to use Blender with any LLM
MaxKB
1Panel-dev🔥 MaxKB is an open-source platform for building enterprise-grade agents. 强大易用的开源企业级智能体平台。
AutoBrowser MCP
autobrowser-aiBrowser MCP is a Model Context Provider (MCP) server that allows AI applications to control your browser
XcodeBuildMCP
cameroncookeA Model Context Protocol (MCP) server and CLI that provides tools for agent use when working on iOS and macOS projects.
コメント