gemini-pr-reviews
Enables fetching and analyzing Gemini Code Assist reviews from GitHub pull requests, with smart defaults and pagination.
README
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
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:
- Go to https://github.com/settings/tokens
- Click "Generate new token (classic)"
- Give it a name and select the
reposcope - Copy 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 format- If just repo name:
"YtbMp3Indir"- uses authenticated user as owner - Full format:
"owner/RepoName"
- If just repo name:
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 2 -
Get ALL reviews (not just after your last comment):
Use get_gemini_reviews for repo YtbMp3Indir with after_last_review false -
Use 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 PR - Fetches 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.py
Standalone Script
You can also use the original CLI script directly:
python3 gemini_pr_reviews.py --repo owner/repo --pr 123
License
MIT
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.