github-mcp-bridge
A lightweight MCP server that exposes GitHub operations as tools over HTTP, enabling any MCP-compatible client to interact with GitHub repositories without a built-in connector.
README
github-mcp-bridge
A lightweight TypeScript MCP (Model Context Protocol) server for GitHub. It exposes GitHub operations as MCP tools over HTTP, so any MCP-compatible client can interact with GitHub repositories without needing a built-in GitHub connector.
The GitHub PAT lives server-side only. Clients authenticate to the bridge using a shared secret.
How it works
MCP Client ──bearer token──▶ github-mcp-bridge ──GitHub PAT──▶ GitHub API
- The client sends a JSON-RPC 2.0
tools/callrequest to the bridge with a bearer token - The bridge validates the token against
CONNECTOR_SECRET - The bridge calls the GitHub API using
GITHUB_PATand returns the result
The bridge also exposes tools/list so any client can discover all available tools and their input schemas at runtime — no manual tool configuration needed.
Available tools
Repositories
| Tool | Description |
|---|---|
list_repositories |
List repositories accessible to the configured PAT |
Branches
| Tool | Description |
|---|---|
list_branches |
List branches for a repository |
get_branch |
Get branch details including latest commit SHA, message, and protection status |
create_branch |
Create a branch from an existing base branch |
Files
| Tool | Description |
|---|---|
get_file_contents |
Get the contents of a file in a repository |
get_multiple_files |
Get the contents of multiple files in a repository |
list_directory |
List files and directories at a repository path |
upsert_file |
Create or update a file in a repository branch |
Pull Requests
| Tool | Description |
|---|---|
list_open_pull_requests |
List open pull requests for a repository |
get_pull_request |
Get a pull request by number |
list_pull_request_files |
List files changed in a pull request |
list_pull_request_comments |
List comments on a pull request |
get_pull_request_diff |
Get the unified diff for a pull request |
create_pull_request |
Create a pull request |
update_pull_request |
Update a pull request |
Issues
| Tool | Description |
|---|---|
list_issues |
List issues for a repository (excludes pull requests) |
get_issue |
Get a single issue by number |
create_issue |
Create a new issue |
update_issue |
Update an existing issue (title, body, state, labels, assignees) |
link_issue_to_pull_request |
Link an issue to a PR using a closing keyword (closes/fixes/resolves) |
list_issue_comments |
List all comments on an issue |
add_issue_comment |
Post a comment on an issue |
Commits
| Tool | Description |
|---|---|
list_commits |
List commits, optionally filtered by branch or file path |
get_commit |
Get full commit detail by SHA or ref, including changed files and diff stats |
Search
| Tool | Description |
|---|---|
search_code |
Search for code within a repository — returns file paths and match fragments |
search_files |
Search for files by name or path pattern using the git tree |
Getting started
Prerequisites
- Node.js >= 24
- A GitHub Personal Access Token with
reposcope (or a fine-grained PAT scoped to the repositories you need)
Local development
# 1. Clone the repo
git clone https://github.com/SamNewhouse/github-mcp-bridge.git
cd github-mcp-bridge
# 2. Install dependencies
npm install
# 3. Set up environment variables
cp .env.example .env
# Edit .env and fill in GITHUB_PAT and CONNECTOR_SECRET
# 4. Start the dev server (hot-reloads on change)
npm run dev
The server starts on http://localhost:3000 by default (configurable via PORT in .env).
Environment variables
| Variable | Required | Description |
|---|---|---|
GITHUB_PAT |
✅ | GitHub Personal Access Token |
CONNECTOR_SECRET |
✅ | Shared secret used to authenticate requests to the bridge |
PORT |
✗ | HTTP port (default: 3000) |
Deploying
The bridge is a standard Node.js HTTP server. It can be deployed anywhere that runs Node.js.
Set the three environment variables (GITHUB_PAT, CONNECTOR_SECRET, and optionally PORT) in your hosting environment, then run:
npm run build
npm start
Once deployed, use the root URL as the MCP endpoint and point your client at it.
Verifying the server
Health check
curl http://localhost:3000/health
# {"ok":true}
Discover all tools
curl -s -X POST http://localhost:3000 \
-H "Authorization: Bearer $CONNECTOR_SECRET" \
-H "Content-Type: application/json" \
-d '{"jsonrpc":"2.0","id":1,"method":"tools/list","params":{}}'
Call a tool
curl -s -X POST http://localhost:3000 \
-H "Authorization: Bearer $CONNECTOR_SECRET" \
-H "Content-Type: application/json" \
-d '{
"jsonrpc": "2.0",
"id": 1,
"method": "tools/call",
"params": {
"name": "list_branches",
"arguments": {
"owner": "your-org",
"repo": "your-repo"
}
}
}'
Connecting an MCP client
Configure your MCP client with:
| Setting | Value |
|---|---|
| URL | Your deployment URL |
| Auth type | Bearer token / API key |
| Secret | Your CONNECTOR_SECRET value |
The client can call tools/list at any time to discover all available tools and their input schemas dynamically.
Scripts
| Command | Description |
|---|---|
npm run dev |
Start dev server with hot-reload |
npm run build |
Compile TypeScript to dist/ |
npm start |
Run compiled server from dist/ |
npm test |
Type-check without emitting |
npm run format |
Format code with Prettier |
Security
- Keep
GITHUB_PATserver-side only — never expose it to clients - Use a fine-grained PAT with the minimum repository permissions needed
- Use a long random string for
CONNECTOR_SECRET(e.g.openssl rand -base64 32) - Rotate
CONNECTOR_SECRETimmediately if it is ever exposed - Rotate
GITHUB_PATimmediately if it is ever exposed - Never log or commit secrets
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。