go-review-mcp
Code reviewing following the Go style guides directly from official and recommended sources.
Go Review MCP
A Model Context Protocol (MCP) server that provides real-time access to official Go style guides for intelligent code review.
Overview
Go Review MCP fetches the latest style guides directly from official and recommended sources:
Official
Recommended
Features
Tools
- search_live_guide - Search through all live-fetched Go style guide content for specific patterns (auto-fetches guides on first use)
- get_guide_topic - Extract topic-specific content from all live guides (naming, errors, concurrency, testing, interfaces, formatting, comments, imports, context)
- get_review_guidelines - Get curated review guidelines for specific topics
Prompts
- review_code - Comprehensive Go code review
- check_naming - Review naming conventions
- check_error_handling - Review error handling patterns
- check_concurrency - Review concurrency patterns
- check_testing - Review test code quality
- check_interfaces - Review interface design
Installation
Using go install
go install github.com/BrunoKrugel/go-review-mcp@latest
From source
git clone https://github.com/BrunoKrugel/go-review-mcp.git
cd go-review-mcp
make build
Configuration
Claude Desktop
Add to your claude_desktop_config.json:
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
{
"mcpServers": {
"go-review": {
"command": "go-review-mcp",
"args": []
}
}
}
If installed from source, use the full path to the binary:
{
"mcpServers": {
"go-review": {
"command": "/path/to/go-review-mcp/bin/go-review-mcp",
"args": []
}
}
}
Cursor or VS Code Extension
{
"mcpServers": {
"go-review": {
"command": "go-review-mcp",
"args": []
}
}
}
Crush
{
"go-review-mcp": {
"type": "stdio",
"command": "go-review-mcp",
"timeout": 120,
"disabled": false
},
}
HTTP Mode (for testing with MCP Inspector)
# Start server in HTTP mode
go-review-mcp --transport http --port 8080
# Or with environment variable
TRANSPORT=http PORT=8080 go-review-mcp
Then use MCP Inspector to test:
npx @modelcontextprotocol/inspector http://localhost:8080/mcp
Usage
Basic Workflow
-
Review your code:
Use the review_code prompt with your Go code -
Get specific guidance:
Use get_guide_topic for focused advice on topics like "naming" or "errors"Guides are automatically fetched on first use and cached for 24 hours.
-
Search for patterns:
Use search_live_guide to find specific content across all guides
Example Interactions
Comprehensive code review:
Please review this Go code:
[paste your code]
Check specific topics:
Check the naming conventions in this code
Search for guidelines:
Search the style guides for information about error wrapping
Get topic-specific content:
Show me all guidelines about concurrency from the style guides
Available Topics
The following topics are automatically indexed across all guides:
naming- Naming conventions and identifier ruleserrors- Error handling and error wrapping patternsconcurrency- Goroutines, channels, mutexes, context usagetesting- Test structure, table-driven tests, best practicesinterfaces- Interface design and usage patternsformatting- Code formatting and stylecomments- Documentation and comment conventionsimports- Import organization and groupingcontext- Context.Context usage and patterns
Configuration Options
The server supports configuration through environment variables:
TRANSPORT- Transport mode:stdio(default) orhttpPORT- HTTP port when using http transport (default: 8080)CACHE_TTL- Cache duration for fetched guides (default: 24h)
Example:
CACHE_TTL=12h go-review-mcp
Development
Building
make build
Running tests
make test
Running locally
make run
How It Works
- Lazy Fetching: Style guides are automatically fetched on first tool use (no manual initialization needed)
- Content Parsing: Guides are parsed into structured sections with automatic topic indexing
- Unified Search: All search and topic queries automatically scan across all fetched guides
- MCP Integration: Simple tools and prompts are exposed via the Model Context Protocol
Key Features
- Auto-Fetch on Demand: Guides are fetched lazily on first use — no manual fetch step required
- Concurrent Fetching: All guides are fetched in parallel for fast initialization
- Smart Caching: 24-hour cache reduces network requests and improves performance
- Topic Indexing: Automatic categorization of content by topics (naming, errors, concurrency, etc.)
- No Configuration: Works out of the box - no need to choose which guides to use
Architecture
┌─────────────────┐
│ AI Assistant │
│ (Claude, etc.) │
└────────┬────────┘
│ MCP Protocol (stdio/http)
┌────────▼──────────────────────┐
│ Go Review MCP Server │
│ - 3 Tools │
│ - 6 Prompts │
│ - Concurrent HTTP Fetcher │
│ - 24h Cache (in-memory) │
│ - Content Parser │
│ - Topic Indexer │
└────────┬──────────────────────┘
│ HTTPS (parallel)
┌────────▼──────────────────────┐
│ Official Style Guide Sources │
│ ├─ Google Go Guide │
│ ├─ Google Go Decisions │
│ ├─ Google Go Best Practices │
│ └─ Uber Go Style Guide │
└───────────────────────────────┘
Data Flow
- On-Demand Fetch: First tool call triggers concurrent fetch of all 4 guides
- Parsing: Each guide is parsed into sections with metadata
- Indexing: Sections are automatically categorized by topics
- Query: Search and topic tools scan all indexed content
- Caching: Results are cached for 24 hours to improve performance
Requirements
- Go 1.21 or later
- Internet connection for fetching live guides (embedded rules work offline)
License
MIT License - see LICENSE file for details
Contributing
Contributions are welcome! Please feel free to submit a Pull Request.
Acknowledgments
This tool aggregates content from:
- Google Go Style Guide - CC-BY-3.0
- Uber Go Style Guide - MIT
संबंधित सर्वर
Kone.vc
प्रायोजकMonetize your AI agent with contextual product recommendations
Mente
Connect your AI assistant to your personal knowledge base. Search, save links, create notes and to-dos. AI processes everything automatically.
Vedit-MCP
Perform basic video editing operations using natural language commands. Requires ffmpeg to be installed.
ClickUp MCP
Integrate ClickUp project management with AI to manage tasks, lists, and spaces.
Online PR - Press Release Distribution
Search PR agencies, browse publications, buy press release distribution & media placements. Zero-config, no API key needed.
ClickUp MCP Server
Apify-hosted MCP server for ClickUp with 20 tools. Tasks, spaces, folders, lists, views, docs, and custom fields. No local setup needed.
Email MCP for Gmail, iCloud and microsoft
Organize, flag, read, delete, and clean email with AI.
Bear Notes
Access and manage your notes from the Bear App.
ClickUp
Interact with the ClickUp API to manage tasks, lists, and spaces, automating project planning and workflows.
Ramp
Interact with Ramp's Developer API to run analysis on your spend and gain insights leveraging LLMs
ClickUp MCP Server (Enhanced Fork)
An MCP server for integrating ClickUp tasks with AI applications, featuring task dependency management and bug fixes.