mennansevim/mcp-server
If you are the rightful owner of mcp-server and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to dayong@mcphub.com.
The MCP Code Review Server is a platform-agnostic, AI-powered code review server that integrates with various development platforms and supports manual code review through MCP tools.
🤖 MCP Code Review Server
Platform-agnostic AI-powered code review server with webhook support and MCP integration.
✨ Features
- 🔌 Platform Agnostic: Single webhook endpoint for GitHub, GitLab, Bitbucket, Azure DevOps
- 🤖 AI-Powered: Uses Groq (Llama 3.3), Claude, or GPT-4 for intelligent code review
- 💬 Multiple Comment Styles: Summary comments, inline comments, or both
- 🎯 Focused Analysis: Security, performance, bugs, code quality
- 🔧 MCP Tools: Manual code review via Claude Desktop or other MCP clients
- 🚀 Easy Integration: Simple pipeline configuration
🏗️ Architecture
Pipeline Webhook → MCP Server → Platform Detection → AI Review → Post Comments
📦 Installation
1. Clone and Setup
cd python
python -m venv venv
source venv/bin/activate # Windows: venv\Scripts\activate
pip install -r requirements.txt
2. Configuration
Copy .env.example to .env and configure:
# AI Provider (choose one)
GROQ_API_KEY=your_key
# or
ANTHROPIC_API_KEY=your_key
# or
OPENAI_API_KEY=your_key
# Platform Tokens
GITHUB_TOKEN=your_github_token
GITLAB_TOKEN=your_gitlab_token
BITBUCKET_USERNAME=your_username
BITBUCKET_APP_PASSWORD=your_password
AZURE_DEVOPS_PAT=your_azure_pat
AZURE_DEVOPS_ORG=https://dev.azure.com/your-org
Edit config.yaml for review preferences:
ai:
provider: "groq" # or "anthropic" or "openai"
model: "llama-3.3-70b-versatile" # Groq models
# model: "claude-3-5-sonnet-20241022" # Anthropic
# model: "gpt-4-turbo-preview" # OpenAI
review:
comment_strategy: "both" # summary, inline, both
report_levels:
- critical
- high
- medium
block_on_critical: true
3. Run Server
python server.py
Server runs on http://localhost:8000
🔧 Pipeline Integration
Bitbucket Pipelines
See examples/bitbucket-pipelines.yml
pipelines:
pull-requests:
'**':
- step:
script:
- curl -X POST $REVIEW_SERVER_URL/webhook ...
GitHub Actions
See examples/github-actions.yml
- name: Trigger AI Review
run: |
curl -X POST ${{ secrets.REVIEW_SERVER_URL }}/webhook ...
GitLab CI/CD
See examples/gitlab-ci.yml
ai-code-review:
script:
- curl -X POST $REVIEW_SERVER_URL/webhook ...
Azure Pipelines
See examples/azure-pipelines.yml
- script: |
curl -X POST $(REVIEW_SERVER_URL)/webhook ...
🎯 MCP Tools (Manual Review)
Use from Claude Desktop or any MCP client:
1. Review Code
{
"tool": "review_code",
"arguments": {
"code": "def login(username, password):\n query = f\"SELECT * FROM users WHERE username='{username}'\"",
"focus": ["security", "bugs"]
}
}
2. Analyze Diff
{
"tool": "analyze_diff",
"arguments": {
"diff": "--- a/file.py\n+++ b/file.py\n..."
}
}
3. Security Scan
{
"tool": "security_scan",
"arguments": {
"code": "your_code_here",
"language": "python"
}
}
📊 Review Output
Summary Comment Example
## 🤖 AI Code Review
**Score:** 7/10 ⚠️
### 📝 Summary
Good code structure but found some security concerns...
### 📊 Issues Found
- Total: **5**
- 🔴 Critical: **1**
- 🟠 High: **2**
- 🟡 Medium: **2**
### ⚠️ Important Issues
#### 🔴 SQL Injection Vulnerability
**Severity:** CRITICAL
**Location:** `auth.py` (Line 42)
Using string concatenation for SQL queries...
**Suggestion:**
> Use parameterized queries...
Inline Comments
Comments posted directly on the problematic code lines.
🔒 Security
- Webhook signature verification
- API token authentication
- Environment-based secrets
- No sensitive data logging
🚀 Deployment
Docker
docker build -t mcp-code-review .
docker run -p 8000:8000 --env-file .env mcp-code-review
Production
- Use reverse proxy (nginx, traefik)
- Enable HTTPS
- Set up logging and monitoring
- Configure rate limiting
🛠️ Development
# Install dev dependencies
pip install -r requirements.txt
# Run tests
pytest
# Format code
black .
📝 API Endpoints
GET /- Health checkPOST /webhook- Universal webhook endpointGET /mcp/sse- MCP Server-Sent Events endpoint
🤝 Contributing
Contributions welcome! Please:
- Fork the repository
- Create a feature branch
- Submit a pull request
📄 License
MIT License
🆘 Support
For issues and questions:
- Open an issue on GitHub
- Check documentation
- Review examples
Made with ❤️ for better code reviews