Github CLI MCP Server
Enables interaction with GitHub by wrapping the gh CLI into two generic tools for executing any gh command and suggesting syntax, providing structured JSON output.
README
Github CLI MCP Server
MCP server that wraps the GitHub CLI (gh) into two generic tools — similar to how aws-api-mcp-server wraps the AWS CLI.
Why?
Most GitHub MCP servers expose 25+ individual tools — one per API operation. That's a lot of tools for something the gh CLI already handles in a single binary.
aws-api-mcp-server solved this for AWS by wrapping the entire AWS CLI into one call_aws tool. This project does the same for GitHub — two tools instead of dozens:
call_gh— run anyghcommandsuggest_gh_commands— get help with syntax
No need for a dedicated tool per operation. The gh CLI already covers repos, issues, PRs, workflows, releases, and more. This just gives MCP clients a way to execute those commands directly.
Why not just use gh through a shell tool?
You can — it works. This server just cleans things up:
- Parses
ghoutput into structured JSON when possible, instead of raw terminal text - Handles timeouts so commands don't hang forever
- Includes a
suggest_gh_commandstool for when you're unsure about syntax - Shows up as a dedicated "github" capability instead of generic shell access
Same gh CLI underneath, just a smoother wrapper for MCP clients to work with.
Tools
call_gh— Execute anyghCLI command and get structured outputsuggest_gh_commands— Get command suggestions from a natural language query
Prerequisites
- GitHub CLI installed and authenticated (
gh auth login) - uv (recommended, for running with
uvx)
Installation
pip install gh-cli-mcp-server
Usage with MCP clients
{
"mcpServers": {
"github": {
"command": "uvx",
"args": ["gh-cli-mcp-server"]
}
}
}
Examples
Just ask your MCP client naturally — it'll use call_gh under the hood:
| You say | What runs |
|---|---|
| "List my repos" | gh repo list --limit 30 |
| "Show open issues on my project" | gh issue list --repo owner/repo --state open |
| "Create a PR from this branch to main" | gh pr create --title '...' --body '...' --base main |
| "What workflows does this repo have?" | gh workflow list --repo owner/repo |
| "Show me the last 5 CI runs" | gh run list --repo owner/repo --limit 5 |
| "Who am I logged in as?" | gh auth status |
If the client isn't sure which command to use, it can call suggest_gh_commands with your prompt to get suggestions first.
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。