Unified Github Integration

Unified Github Integration is an MCP server that provides comprehensive access to the GitHub platform API, enabling AI assistants to manage repositories, issues, pull requests, code reviews, and collaborative development workflows. This MCP bridge connects language models to your GitHub account with full API coverage, allowing conversational repository management, automated issue triage, pull request review assistance, and codebase exploration across organizations and personal repositories.

When to Use This MCP Server

Connect this server when...

• You need AI assistance managing GitHub repositories, issues, pull requests, and code reviews through natural language
• Your development workflow is centered on GitHub and you want conversational access to repository data and collaboration
• You are building automation workflows that create issues, label pull requests, or manage GitHub project boards
• You need to search across multiple repositories for code patterns, issues, or pull request discussions
• You want AI-assisted code review that reads pull request diffs and generates review comments

Consider alternatives when...

• Your repositories are hosted on GitLab or Bitbucket, which have their own dedicated MCP servers
• You only need local git operations without interacting with the GitHub API
• You need only a subset of GitHub features that a more specialized tool might handle

Quick Start

Copy
# .mcp.json configuration
{
  "mcpServers": {
    "github": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-github"],
      "env": {
        "GITHUB_PERSONAL_ACCESS_TOKEN": "ghp_your_token_here"
      }
    }
  }
}

Connection setup:

1. Generate a Personal Access Token at github.com/settings/tokens
2. Select the required scopes: repo, read:org, read:user (add more as needed)
3. Add the configuration above to your .mcp.json file with your token
4. Restart your MCP client to connect to GitHub

Example tool usage:

# List repository issues
> Show me all open issues labeled "bug" in the frontend-app repository

# Review a pull request
> Summarize the changes in PR #156 and highlight any potential issues

# Search code
> Find all usages of the deprecated fetchUser function across our organization's repos

Core Concepts

Concept	Purpose	Details
Repositories	Code hosting	Git repositories with branches, commits, releases, and collaboration features
Issues	Work tracking	Bug reports, feature requests, and task tracking with labels, assignees, and milestones
Pull Requests	Code review	Proposed code changes with diffs, review comments, approval workflows, and merge controls
Organizations	Team grouping	GitHub organizations containing repositories, teams, and org-level settings
Personal Access Token	API authentication	Scoped token granting API access to your GitHub account resources

Architecture:

+------------------+       +------------------+       +------------------+
|  GitHub          |       |  GitHub MCP      |       |  AI Assistant    |
|  API             |<----->|  Server (npx)    |<----->|  (Claude, etc.)  |
|  api.github.com  | HTTPS |  stdio transport  | stdio |                  |
+------------------+       +------------------+       +------------------+
        |
        v
+------------------------------------------------------+
|  Repos > Issues > PRs > Reviews > Code > Actions      |
+------------------------------------------------------+

Configuration

Parameter	Type	Default	Description
GITHUB_PERSONAL_ACCESS_TOKEN	string	(required)	Personal Access Token with appropriate scopes for your workflow
default_owner	string	none	Default repository owner (user or org) for operations when not specified
default_repo	string	none	Default repository name to scope operations when not specified
per_page	integer	30	Number of results per page for list operations (max 100)
api_url	string	https://api.github.com	GitHub API base URL (change for GitHub Enterprise Server)

Best Practices

1. Use fine-grained Personal Access Tokens. GitHub fine-grained tokens can be scoped to specific repositories and permissions. Use these instead of classic tokens to limit the MCP server's access to only the repositories and actions your workflow requires.

2. Set default owner and repo for focused workflows. If you primarily work with a single repository, configure defaults to avoid specifying them in every request. This streamlines conversations and reduces the chance of operating on the wrong repository.

3. Leverage labels and milestones for effective issue queries. Maintain a consistent labeling scheme across repositories. This enables the AI to filter and prioritize issues through natural language queries like "show me high-priority bugs assigned to me."

4. Use the AI for pull request review preparation. Before manually reviewing a PR, ask the AI to summarize changes, identify affected components, and flag potential issues. This pre-review analysis helps focus attention on the most critical parts of the change.

5. Monitor API rate limits during intensive sessions. GitHub enforces 5,000 requests per hour for authenticated requests. During intensive sessions, monitor remaining allowance to avoid hitting limits that block further operations.

Common Issues

"Bad credentials" error when connecting. Verify your Personal Access Token is valid and not expired. Classic tokens do not expire by default, but fine-grained tokens may have an expiration date. Generate a new token if the current one is invalid.

Repository operations return "Not Found" for private repos. Ensure your token has the repo scope for private repository access. Tokens with only public_repo scope cannot access private data.

Rate limiting during bulk operations. GitHub's 5,000 requests/hour limit can be exhausted during intensive operations. Use pagination wisely and avoid unnecessary repeated requests for the same data.