diff-explainer
AI-powered git diff analysis with human-readable explanations, risk flags, and review checklists. Enables to explain any diff text or currently staged git changes through an MCP server.
README
diff-explainer
AI-powered git diff analysis: human-readable explanations, risk flags, and review checklists.
Works as a CLI, Python library, and MCP server tool.
Setup
pip install -r requirements.txt
# Optional: set for LLM-powered summaries
export OPENAI_API_KEY="sk-..."
Risk detection works without an API key. Only summaries and review checklists require the LLM.
Usage
CLI
# Pipe from git diff
git diff | python cli.py
# Read from file
python cli.py --file examples/sample.diff
# Auto-run git diff in current repo
python cli.py --git
# JSON output
python cli.py --file examples/sample.diff --format json
Python Library
from explainer import explain_diff
with open("changes.diff") as f:
diff_text = f.read()
result = explain_diff(diff_text)
print(result.summary)
print(result.risk_flags)
print(result.files_changed)
MCP Server (opencode integration)
Add to your opencode.json:
{
"mcpServers": {
"diff-explainer": {
"type": "stdio",
"command": "python",
"args": ["server.py"],
"cwd": "/path/to/diff-explainer"
}
}
}
Available tools:
explain_diff_tool(diff_text)— Explain any diff textexplain_staged()— Explain currently staged git changes
Run standalone: python server.py
Features
- Hybrid analysis: Regex risk detection (offline) + LLM summaries (optional)
- 7 risk patterns: Removed error handling, changed timeouts, auth modifications, missing env defaults, large changes, tech debt markers, hardcoded secrets
- Graceful degradation: Works without API key — risk flags and file stats always available
- Multiple interfaces: CLI, library import, MCP server
Output Example
# Diff Explanation
## Summary
This change reduces retry resilience in the auth service and introduces a hardcoded API key.
**Category:** config_change
## Files Changed
- `src/auth.py` — +3/-7 lines
- `src/config.py` — +5/-1 lines
## Risk Flags
- 🔴 **HIGH**: Removed error handling code
- 🔴 **HIGH**: Modified auth/permission checks
- 🔴 **HIGH**: Possible hardcoded secret
- 🟡 **MEDIUM**: Changed retry/timeout values
- 🟢 **LOW**: Added TODO/FIXME/HACK comment
## Review Checklist
- [ ] Is removing the try/except intentional? What handles AuthError now?
- [ ] Why were retries reduced from 3 to 1?
- [ ] The hardcoded API key should be moved to a secrets manager.
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.