LeeSinLiang

CommitLM — AI skill for Claude Code

AI community

Teach your LLM coding agents what just happened: Auto docs and commit message generation, with LLM.

How to install CommitLM

This entry records only its repository, not the path inside it, so there is no exact command to give. Open LeeSinLiang/CommitLM and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What CommitLM does

Teach your LLM coding agents what just happened: Auto docs and commit message generation, with LLM.

Alternatives in AI

  • Gsap Skills — Official AI skills for GSAP 1.7k ★
  • Claudekit — by Carl Rannaberg - Impressive CLI toolkit providing auto-save checkpointing, code quality hooks, specificatio 633 ★
  • Comfyui Workflow Skill — Natural language → ComfyUI workflow JSON 180 ★

README

CommitLM — Git AI Companion That Teaches AI and Your Team What Just Happened

[![PyPI version](https://img.shields.io/pypi/v/commitlm.svg)](https://pypi.org/project/commitlm/) [![Python Versions](https://img.shields.io/pypi/pyversions/commitlm.svg)](https://pypi.org/project/commitlm/) [![License](https://img.shields.io/badge/License-Apache%202.0-blue.svg)](https://opensource.org/licenses/Apache-2.0)

**Automated Documentation and Commit Message Generation for Every Git Commit**

AI coding agents are powerful but stateless—they lack context on why a change was made. This leads to guesswork and bugs.

While most "AI commit" tools stop at the message, CommitLM is an AI-native git tool that creates a tiny, structured docs per commit. This acts as a **briefing for LLM coding agents** (like Copilot, Gemini CLI, Claude, etc.), so they can:

  • ✅ Pick up exactly where you or another agent left off.
  • 🧠 Respect constraints and project-specific nuances.
  • 🔄 Update callers, tests, and migrations with fewer misses.
pip install commitlm
commitlm init

Why CommitLM?

  • 🚀 Rapid Documentation for AI Agents: Eliminate manual documentation and commit message writing, while ensuring your AI agents pick up where you left off.
  • ⚡ Zero Friction: Works automatically via Git hooks - no workflow changes needed
  • 📝 Memory-savvy & long-context: 8-bit quantization and YaRN for extended context lengths
  • 🤖 Flexible AI: Choose from multiple LLM providers or run models locally
  • 📝 Create a Living Knowledge Base: Your repo becomes a self-updating source of truth, making onboarding and handoffs seamless for both humans and AI.
  • 🔒 Privacy First: Run local models for complete data privacy

Table of Contents