GitHub Actions MCP server

Enables management of GitHub Actions workflows with tools for listing, viewing, triggering, canceling, and rerunning workflows, along with detailed run and job information for CI/CD pipeline monitoring and troubleshooting.
Back to servers
Provider
ko1ynnky
Release date
Mar 21, 2025
Language
TypeScript
Stats
32 stars

The GitHub Actions MCP Server provides an interface for AI assistants to interact with GitHub Actions workflows through a Model Context Protocol (MCP) server. It enables complete workflow management, detailed analysis, and secure operations across multiple AI coding platforms.

Installation

Prerequisites

  • Node.js (v14 or higher)
  • npm (v6 or higher)
  • A GitHub Personal Access Token with appropriate permissions

Installation on Unix/Linux/macOS

Clone the repository and build the server:

git clone https://github.com/ko1ynnky/github-actions-mcp-server.git
cd github-actions-mcp-server
npm install
npm run build

Installation on Windows

On Windows systems, you can use either of these methods:

git clone https://github.com/ko1ynnky/github-actions-mcp-server.git
cd github-actions-mcp-server
npm install
npm run build:win

Or use the included batch file:

run-server.bat [optional-github-token]

Configuration

Setting Up Your GitHub Token

For full functionality, you need to set your GitHub Personal Access Token:

On Unix/Linux/macOS:

export GITHUB_PERSONAL_ACCESS_TOKEN=your_github_token_here

On Windows:

set GITHUB_PERSONAL_ACCESS_TOKEN=your_github_token_here

Or pass it as a parameter to the Windows batch file:

run-server.bat your_github_token_here

Integrating with AI Assistants

Claude Desktop

Add the following to your claude_desktop_config.json:

{
  "mcpServers": {
    "github-actions": {
      "command": "node",
      "args": [
        "<path-to-mcp-server>/dist/index.js"
      ],
      "env": {
        "GITHUB_PERSONAL_ACCESS_TOKEN": "<YOUR_TOKEN>"
      }
    }
  }
}

Codeium and Windsurf

Add this configuration to your MCP config file (typically at ~/.codeium/windsurf/mcp_config.json on Unix-based systems or %USERPROFILE%\.codeium\windsurf\mcp_config.json on Windows):

{
  "mcpServers": {
    "github-actions": {
      "command": "node",
      "args": [
        "<path-to-mcp-server>/dist/index.js"
      ],
      "env": {
        "GITHUB_PERSONAL_ACCESS_TOKEN": "<YOUR_TOKEN>"
      }
    }
  }
}

Using the MCP Server

Available Tools

Listing Workflows

const result = await listWorkflows({
  owner: "your-username",
  repo: "your-repository",
  page: 1,        // Optional
  perPage: 30     // Optional
});

Getting Workflow Details

const result = await getWorkflow({
  owner: "your-username",
  repo: "your-repository",
  workflowId: "ci.yml" // Can be ID number or filename
});

Checking Workflow Usage

const result = await getWorkflowUsage({
  owner: "your-username",
  repo: "your-repository",
  workflowId: "ci.yml"
});

Listing Workflow Runs

const result = await listWorkflowRuns({
  owner: "your-username",
  repo: "your-repository",
  workflowId: "ci.yml",  // Optional
  actor: "username",     // Optional
  branch: "main",        // Optional
  status: "completed",   // Optional
  event: "push"          // Optional
});

Getting Workflow Run Details

const result = await getWorkflowRun({
  owner: "your-username",
  repo: "your-repository",
  runId: 123456789
});

Getting Jobs for a Run

const result = await getWorkflowRunJobs({
  owner: "your-username",
  repo: "your-repository",
  runId: 123456789,
  filter: "latest"  // Optional: 'latest' or 'all'
});

Triggering a Workflow

const result = await triggerWorkflow({
  owner: "your-username",
  repo: "your-repository",
  workflowId: "ci.yml",
  ref: "main",
  inputs: {
    environment: "production",
    debug: "true"
  }
});

Canceling a Workflow Run

const result = await cancelWorkflowRun({
  owner: "your-username",
  repo: "your-repository",
  runId: 123456789
});

Re-running a Workflow

const result = await rerunWorkflow({
  owner: "your-username",
  repo: "your-repository",
  runId: 123456789
});

Troubleshooting

Common Issues

  1. Authentication Errors

    • Verify your GitHub token has the correct permissions
    • Make sure the token is correctly set as an environment variable
  2. Rate Limiting

    • If you encounter rate limit errors, reduce the frequency of requests
    • The server implements rate limiting to stay within GitHub API limits
  3. Type Validation Errors

    • The server implements flexible validation to handle most GitHub API response variations
    • If you encounter persistent errors, they may be due to API changes
  4. Windows-Specific Issues

    • If encountering build errors, ensure TypeScript is installed: npm install -g typescript
    • For permission issues, run the command prompt with appropriate permissions
    • Verify your Node.js version compatibility with node --version

How to add this MCP server to Cursor

There are two ways to add an MCP server to Cursor. The most common way is to add the server globally in the ~/.cursor/mcp.json file so that it is available in all of your projects.

If you only need the server in a single project, you can add it to the project instead by creating or adding it to the .cursor/mcp.json file.

Adding an MCP server to Cursor globally

To add a global MCP server go to Cursor Settings > MCP and click "Add new global MCP server".

When you click that button the ~/.cursor/mcp.json file will be opened and you can add your server like this:

{
    "mcpServers": {
        "cursor-rules-mcp": {
            "command": "npx",
            "args": [
                "-y",
                "cursor-rules-mcp"
            ]
        }
    }
}

Adding an MCP server to a project

To add an MCP server to a project you can create a new .cursor/mcp.json file or add it to the existing one. This will look exactly the same as the global MCP server example above.

How to use the MCP server

Once the server is installed, you might need to head back to Settings > MCP and click the refresh button.

The Cursor agent will then be able to see the available tools the added MCP server has available and will call them when it needs to.

You can also explictly ask the agent to use the tool by mentioning the tool name and describing what the function does.

Want to 10x your AI skills?

Get a free account and learn to code + market your apps using AI (with or without vibes!).

Nah, maybe later