gemini-pr-reviews
Fetches Gemini Code Assist reviews from GitHub pull requests, providing tools to retrieve and analyze code reviews from the Gemini Code Assist bot.
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@gemini-pr-reviewsget Gemini reviews for repo my-org/my-repo"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Gemini PR Reviews MCP Server
MCP (Model Context Protocol) server for fetching Gemini Code Assist reviews from GitHub pull requests. This tool allows Claude Desktop to retrieve and analyze code reviews from Gemini Code Assist bot on GitHub PRs.
Features
Fetch all Gemini Code Assist reviews from a GitHub PR
Get reviews after your last
/gemini reviewcomment (default behavior)Smart defaults: auto-detect repository owner and last PR
GitHub token authentication support
Full pagination support for large PRs
Retrieves all comment types (reviews, line comments, issue comments)
Raw JSON output for maximum flexibility
Easy installation with provided script
Related MCP server: GitHub Code Review Assistant
Installation
Clone or download this repository
Run the installation script:
cd /path/to/gh_tool ./install.shThis will:
Create a Python virtual environment
Install all required dependencies
Create a
.envfile (if it doesn't exist)Make scripts executable
Edit the
.envfile with your GitHub token:GITHUB_TOKEN=your_actual_github_token_hereTo get a GitHub token:
Click "Generate new token (classic)"
Give it a name and select the
reposcopeCopy the generated token
Configuration for Claude Desktop
Add this server to your Claude Desktop configuration (~/Library/Application Support/Claude/claude_desktop_config.json):
{
"mcpServers": {
"gemini-pr-reviews": {
"command": "/path/to/gh_tool/run.sh"
}
}
}Replace /path/to/gh_tool with the actual path to this directory.
Note: The run.sh script automatically activates the virtual environment and runs the server. Your GitHub token should be in the .env file.
Available Tool
get_gemini_reviews
Get Gemini Code Assist reviews from a GitHub PR. Can fetch all reviews or only those after your last '/gemini review' comment.
Parameters:
repo(string, required): Repository name or owner/repo formatIf just repo name:
"YtbMp3Indir"- uses authenticated user as ownerFull format:
"owner/RepoName"
pr(integer, optional): PR number (uses last PR if not specified)after_last_review(boolean, optional): If true, only fetch reviews after your last '/gemini review' comment (default: true)username(string, optional): GitHub username (only used when after_last_review is true, defaults to authenticated user)
Note: The tool returns raw JSON data for maximum flexibility. By default, it fetches reviews after your last '/gemini review' comment.
Usage Examples
Simplest usage (most common) - Get reviews after your last comment:
Use get_gemini_reviews for repo YtbMp3IndirThis automatically:
Uses your GitHub username as owner
Finds the last PR
Gets reviews after your last '/gemini review' comment
Specify a PR number:
Use get_gemini_reviews for repo YtbMp3Indir pr 2Get ALL reviews (not just after your last comment):
Use get_gemini_reviews for repo YtbMp3Indir with after_last_review falseUse full repo path (if needed):
Use get_gemini_reviews for repo someoneelse/TheirRepo
How It Works
The server fetches Gemini Code Assist comments from GitHub PRs using the GitHub API v3. It supports three types of comments:
Reviews: General PR reviews with overall feedback
Line Comments: Code-specific comments on particular lines
Issue Comments: General discussion comments on the PR
When after_last_review is true (default), the tool:
Finds your last
/gemini reviewcomment in the PRFetches all Gemini bot responses after that timestamp
Returns them sorted by type and date
Notes
The server requires Python 3.7+
GitHub API rate limits apply (5000 requests/hour with token, 60 without)
Authentication via GitHub token is highly recommended
The server returns raw JSON data for maximum flexibility
All API calls use pagination to handle large PRs
Supports typo variants like "/genimi review"
Testing
Run the test script to verify the server is working:
python3 test_raw_output.pyStandalone Script
You can also use the original CLI script directly:
python3 gemini_pr_reviews.py --repo owner/repo --pr 123License
MIT
This server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Latest Blog Posts
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/mytsx/mcp-gemini-pr-reviews'
If you have feedback or need assistance with the MCP directory API, please join our Discord server