Gooseworks MCP
A Model Context Protocol (MCP) server that exposes GitHub tools for repository management, code search, issue creation, and pull request operations.
README
Gooseworks MCP
An Express server that exposes tools through the Model Context Protocol (MCP) Streamable HTTP transport, including a set of GitHub tools.
The MCP endpoint is:
http://localhost:4000/mcp
Every request must include the bearer token configured by MCP_AUTH_TOKEN.
Requirements
- Node.js
v25.2.1 - pnpm
11.5.3 - Docker Engine and Docker Compose v2 for container deployment
- A GitHub personal access token with the permissions required by the tools you plan to use
Configuration
The server loads variables from a local .env file or the process environment:
| Variable | Required | Default | Description |
|---|---|---|---|
PORT |
No | 4000 |
Port on which the HTTP server listens |
GITHUB_TOKEN |
Yes | None | GitHub personal access token used by Octokit |
MCP_AUTH_TOKEN |
Yes | None | Bearer token required for every MCP request |
Create a local environment file from the template:
cp .env.template .env
Edit .env and replace both placeholder token values. Use a long, randomly generated value for MCP_AUTH_TOKEN, and grant the GitHub token only the permissions needed by this server. Never commit .env or paste either token into a public issue, log, or image.
Run Locally
-
Install dependencies:
pnpm install -
Create and configure
.envas described above. -
Start the server:
pnpm startThis runs the same entrypoint as
pnpm exec tsx src/index.ts.
For development with automatic restart after source changes, use:
pnpm dev
To check the TypeScript project without emitting files:
pnpm typecheck
The server is ready when the console reports that it is active on http://localhost:<PORT>/mcp. Configure your MCP client to use that URL and send:
Authorization: Bearer <MCP_AUTH_TOKEN>
Deploy With Docker Compose
-
Install Docker Engine with Docker Compose v2 on the host.
-
Clone this repository and enter its directory:
git clone <repository-url> cd gooseworks-mcp -
Create the deployment environment file:
cp .env.template .env -
Set
GITHUB_TOKENand a strong, privateMCP_AUTH_TOKENin.env. SetPORTif the service should listen on a different host/container port. The same value is used for both sides of the Compose port mapping. -
Build and start the service in the background:
docker compose up --build -d -
Follow the service logs or stop it when needed:
docker compose logs -f gooseworks-mcp docker compose down
The Compose configuration passes the three supported variables into the container and fails early if either required token is missing. For a public deployment, place the service behind an HTTPS reverse proxy and keep port 4000 or your chosen PORT private when possible. MCP clients should use the proxy's HTTPS /mcp URL and the configured bearer token.
Available Tools
The server registers the following tools:
GitHub
- Reading the authenticated profile
- Listing repositories, branches, and commits
- Searching code
- Reading file contents
- Creating issues
- Creating pull requests
License
This project is licensed under the MIT License. It is free to use, fork, modify, and redistribute for any purpose, subject to the license terms.
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。