広告掲載
MCP.so
ログイン
G

Graphify Ts MCP Server

@mohanagy

リポジトリ

基本情報

カテゴリ

その他

トランスポート

stdio

作者

mohanagy

投稿者

Mohammed Naji

Graphify Ts MCP Server について

Local MCP server and context compiler that turns TypeScript/Node workspaces into compact, verifiable context packs for AI coding agents.

Graphify Ts MCP Server の設定

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": {
    "graphify-ts": {
      "command": "npx",
      "args": [
        "-y",
        "@mohammednagy/graphify-ts@latest",
        "serve",
        "graphify-out/graph.json"
      ],
      "env": {
        "GRAPHIFY_TOOL_PROFILE": "core"
      }
    }
  }
}

Graphify Ts MCP Server のツール

ツール一覧はまだありません

この掲載には表示できるツール一覧がまだありません。サーバーにツールがないという意味ではありません。対応機能はプロジェクトのドキュメントをご確認ください。

上のボタンで一時的なサンドボックスでこのサーバーを実行し、ツール一覧を取得できます。

概要

What is Graphify Ts?

Graphify Ts indexes a TypeScript/Node workspace (and PR diffs) into a local knowledge graph and compiles the graph into compact, verifiable context packs for AI coding agents. It runs entirely on your machine with no cloud dependencies and no API keys.

How to use Graphify Ts?

Install globally with npm install -g @mohammednagy/graphify-ts. Run graphify-ts generate . to build the knowledge graph, then graphify-ts claude install (or the equivalent command for Cursor, Copilot CLI, Gemini CLI, Aider, Codex CLI, OpenCode) to wire the agent via MCP or a context-pack-first profile. Use graphify-ts pack or graphify-ts prompt to produce a provider-ready prompt without MCP.

Key features of Graphify Ts

  • Local-first: no cloud required, your code never leaves your laptop.
  • Reduces per-session input tokens by up to 5.28× on measured benchmarks.
  • Supports PR review via pr_impact and review-compare tools.
  • Works with Claude Code, Cursor, Copilot CLI, Gemini CLI, Aider, Codex CLI, and OpenCode.
  • Uses tree-sitter AST for TypeScript/JavaScript with framework-aware passes for 9 substrates.
  • Offers MCP tools like retrieve, impact, call_chain, and community_overview.

Use cases of Graphify Ts

  • Cut token usage and latency on codebase questions by providing pre-compiled context to AI agents.
  • Review pull request diffs with structured blast-radius analysis and ranked review risks.
  • Explore unfamiliar codebases through architecture summaries, call chains, and dependency graphs.
  • Run multi-agent workflows across Claude Code, Codex, Copilot, Cursor, and Gemini with a single graph.

よくある質問

1

What does Graphify Ts do that alternatives do not?

It builds a local knowledge graph once and compiles compact context packs on demand, so agents do not re-discover the codebase from scratch every session. Measured benchmarks show up to 5.28× fewer input tokens and 2.21× fewer tool-call turns.

2

What runtime dependencies are required?

Node.js 20 or later and npm. No API keys, no cloud services, and no external databases are required. The tool is installed as a single global npm package.

3

Where does your code data live?

Everything runs locally on your machine. Your code never leaves the laptop unless you explicitly invoke a model through a connected AI agent.

4

Which languages and frameworks does Graphify Ts support?

Deepest extraction is for TypeScript/JavaScript with framework-aware passes for Express, NestJS, Next.js, React Router, Redux Toolkit, Hono, Fastify, tRPC, and Prisma. Python, Ruby, Go, Java, and Rust use tree-sitter AST. C, Kotlin, C#, Scala, PHP, Swift, and Zig use a generic structural extractor.

5

How does Graphify Ts connect to AI agents?

It provides MCP integration for Claude Code, Cursor, Copilot CLI, and Gemini CLI. For Aider, Codex CLI, and OpenCode, it writes a context-pack-first profile. For Windsurf and others, you pipe the output of `graphify-ts prompt` directly to the agent.

コメント

「その他」の他のコンテンツ

Figs, Relationship Expert

Figs O'Sullivan, LMFT

Couples therapist Figs O'Sullivan (LMFT, 17 years, endorsed by Sue Johnson, creator of EFT) inside your AI. Ask relationship questions answered from his clinical writing, name the pattern in a fight, get a three-step rep

Blender

ahujasid

Open-source MCP to use Blender with any LLM

23.6K

DeepMark

DeepMark helps teachers deliver rapid, consistent marking with meaningful feedback for every student — in a fraction of the time. What once took a week, now takes one free period.

Reactive Resume

amruthpillai

A one-of-a-kind resume builder that keeps your privacy in mind. Completely secure, customizable, portable, open-source and free forever. Try it out today!

39.5K

Hindsight

vectorize-io

Hindsight: Agent Memory That Learns

50.4K
広告掲載