git-commit-messages

guillempuche's avatarfrom guillempuche

Generates consistent git commit messages following project conventions. Use when committing changes, creating PRs, or when asked to write commit messages.

⭐0stars🔀0forks📁View on GitHub🕐Updated Dec 25, 2025

When & Why to Use This Skill

This Claude skill automates the creation of standardized Git commit messages based on specific project conventions. By enforcing rules such as imperative mood subjects, scope derivation from file paths, and categorized change types (feat, fix, refactor, etc.), it ensures a clean, readable, and professional version control history that enhances team collaboration and repository maintainability.

Use Cases

  • Standardizing Version Control: Automatically generate consistent commit messages that adhere to strict team formatting rules, reducing manual effort and cognitive load.
  • Streamlining Pull Requests: Create clear and structured summaries of code changes to facilitate faster peer reviews and better project documentation.
  • Managing Multi-Scope Refactoring: Intelligently handle complex changes across multiple directories by automatically deriving or omitting scopes based on file paths.
  • AI Development Tracking: Use specialized 'ai' tags to distinguish changes in prompts, skills, and agent configurations from traditional software engineering tasks.
  • Automated Change Logging: Ensure every commit includes a past-tense body description, making it easier to generate automated changelogs and release notes.
namegit-commit-messages
descriptionGenerates consistent git commit messages following project conventions. Use when committing changes, creating PRs, or when asked to write commit messages.

Git Commit Message Generator

Format

type(scope): subject in imperative mood

- Body bullet in past tense with period.
- Another change description.

Types

Type When to Use
feat Added new functionality
fix Fixed a bug
refactor Restructured code, no behavior change
chore Dependencies, tooling, configs
docs Documentation
test Tests
cicd CI/CD pipelines, deployment
ai AI/Claude configurations

Rules

  1. Subject: Imperative mood, lowercase after colon, no period, max 72 chars
  2. Scope: Derived from path. When changes span multiple scopes, omit the scope entirely
    • authz – Authorization stack and FGA models
    • infra – Kraftfiles, Dockerfiles, deployment configs
    • nix – Flake and Nix configuration
  3. Body: Past tense, capital start, period at end
  4. No attribution: Never include "Co-Authored-By", "Generated with", or any AI/author attribution
  5. AI-only changes: When changes are exclusively AI-related (skills, prompts, Claude configs), always use ai type—never refactor, chore, or other types
  6. Preview before commit: Always show the proposed commit message to the user for confirmation before executing the commit

Examples

feat(authz): add task permissions with list inheritance
refactor(authz): split monolithic model into modules

- Separated projects.fga and tasks.fga into distinct files.
- Created fga.mod manifest to declare included modules.
chore(nix): update flake inputs to latest versions

- Bumped nixpkgs to 2025-12-21.
- Updated unikraft-nur to v0.12.5.
feat(infra): add Caddy reverse proxy for Unikraft deployment

- Added Kraftfile and rootfs for Caddy unikernel.
- Configured TLS termination and playground proxy.
docs: restructure README for improved clarity
chore: update dependencies and documentation

- Bumped openfga-cli to v0.7.8.
- Updated AGENTS.md with README references.
ai: secure Claude settings by restricting dangerous permissions