GitHub

A Model Context Protocol (MCP) application for automated GitHub PR analysis and issue management. Enables LLMs to fetch PR details, analyse diffs, manage issues, and handle releases through a standardised interface
  • python

5

GitHub Stars

python

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

The MCP for GitHub PR, Issues, Tags and Releases is a bridge that lets your LLM ecosystem automate GitHub repository management. It fetches PR details and diffs, creates and updates issues, and handles tags and releases, all through a standard MCP interface. This enables streamlined PR analysis, issue tracking, and release workflows directly from your preferred LLM tooling.

How to use

You connect to the MCP server from an MCP client or IDE extension. Choose one of two operation modes: stdio (local server) or http (remote/server-accessible). In stdio mode, you run the MCP server locally and interact via standard I/O. In http mode, you start the server and interact over HTTP at a local endpoint.

Prerequisites you need: a GitHub Personal Access Token with repo scope and Python 3.11 or newer. You will set the token as an environment variable when starting the server. Then you launch the MCP server in your preferred mode and point your client to it.

If you run in stdio mode, your client will start the server with a command like uvx and will supply the repository path or current directory as an argument. If you run in http mode, you will enable remote access before starting, and you will connect to the HTTP URL (for example, http://localhost:8080/mcp). Once running, you can perform automated PR analyses, fetch PR details, create and update issues, add comments, and manage releases and tags directly from your LLM prompts.

How to install

Prerequisites you need to install or have available before starting: Python 3.11+, Git, and a working network connection to reach GitHub. You also need a GitHub Personal Access Token with the repo scope.

Step 1: Clone the MCP project repository.

git clone https://github.com/saidsef/mcp-github-pr-issue-analyser.git
cd mcp-github-pr-issue-analyser

Step 2: Start the server in stdio mode (local) with your GitHub token available as an environment variable. This mode runs the MCP server locally and communicates via stdio.

export GITHUB_TOKEN="<github-token>"
uvx ./

Step 3: Alternatively, start the server in http mode (remote access). Enable remote mode by setting the environment variable and then launch the server. You will access it via an HTTP URL.

export GITHUB_TOKEN="<github-token>"
export MCP_ENABLE_REMOTE=true
uvx ./

You can access it via http(s)://localhost:8080/mcp

## Additional content and notes

IDE and LLM integration examples show how to configure clients to connect to the MCP server. You can supply the GitHub token via environment variables and provide the server location you are connecting to.

Environment variables you'll commonly use are the GitHub token and the flag that enables remote access when using the HTTP interface.

## Available tools

### PR analysis

Analyses pull requests and fetch diffs to provide insights on changes, authorship, and context.

### Fetch PR content

Retrieves PR title, description, author, timestamps, and state for an overview.

### Update PRs

Updates PR title or body to reflect changes or clarifications.

### Comment on PRs

Adds general comments to a PR thread to communicate feedback.

### Inline review comments

Adds inline comments on specific lines in PR files for code review.

### Issue management

Creates and updates GitHub issues with titles, bodies, labels, and states.

### Release tagging

Creates tags and publishes releases with changelogs.

### IP information

Retrieves public IPv4 and IPv6 address information.

### Open PR/Issue listing

Lists all open PRs or issues for a user or organization.
Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational
GitHub MCP Server - saidsef/mcp-github-pr-issue-analyser | VeilStrat