MCP.so
ログイン

Unleash MCP Server

@cuongtl1992

基本情報

カテゴリ

その他

ライセンス

MIT

トランスポート

stdio

作者

cuongtl1992

投稿者

Tran Le Cuong

Unleash MCP Server について

A Model Context Protocol (MCP) server implementation that integrates with Unleash Feature Toggle system.

Unleash 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": {
    "unleash-mcp": {
      "command": "npx",
      "args": [
        "@modelcontextprotocol/inspector",
        "node",
        "dist/index.js"
      ]
    }
  }
}

Unleash 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 Unleash MCP Server?

A Model Context Protocol (MCP) server that bridges LLM applications with the Unleash feature flag system. It lets AI applications check feature flag status, expose flag information, create and update flags, and list all projects. Designed for developers who want to integrate Unleash flag management into LLM workflows.

How to use Unleash MCP Server?

Install dependencies with npm i, build with npm run build, then run with npm start. For Claude or Cursor, add an MCP config using npx -y unleash-mcp and set the environment variables UNLEASH_URL, UNLEASH_API_TOKEN, MCP_TRANSPORT, and MCP_HTTP_PORT. The server supports both stdio and HTTP/SSE transports.

Key features of Unleash MCP Server

  • Check feature flag status from Unleash
  • Expose feature flag information to LLMs
  • Create and update feature flags
  • List all projects via MCP resources
  • Supports stdio and HTTP/SSE transport modes

Use cases of Unleash MCP Server

  • An AI assistant queries flag status to decide behavior
  • Automated creation or updating of feature flags through an LLM
  • Listing all projects to provide context to an LLM

よくある質問

1

What are the requirements?

Node.js v18 or higher, TypeScript v5.0 or higher, and access to an Unleash server instance.

2

How do I configure the server?

Set the environment variables `UNLEASH_URL` (your Unleash endpoint), `UNLEASH_API_TOKEN` (API token), `MCP_TRANSPORT` (e.g., `stdio`), and `MCP_HTTP_PORT` (e.g., `3001`).

3

What transports does the server support?

It supports stdio transport and HTTP/SSE transport, with implementations in `src/transport/stdio.ts` and `src/transport/http.ts`.

4

Can I inspect the MCP server during development?

Yes. Use `npm run build` then `npx @modelcontextprotocol/inspector node dist/index.js` for stdio, or run `npm start` then `npx @modelcontextprotocol/inspector` for SSE.

5

Which operations are available?

The server provides tools to get a feature flag, get all projects, and create/update flags; resources expose flags and projects; prompts include single and batch flag checking.

コメント

「その他」の他のコンテンツ