ReddRadar
Agent skill
git-commit
Write clear git commits with Conventional Commits format. Detects project conventions from history and config. Guides commit granularity. Use when: (1) Completing working code, (2) Code builds and tests pass, (3) Ready to save, (4) Before pushing, (5) After review feedback. Triggers: automatically when finishing commitable work that builds and passes tests.
Install this agent skill to your Project
npx add-skill https://github.com/LukasStrickler/ai-dev-atelier/tree/main/content/skills/git-commit
Metadata
Additional technical details for this skill
- author
- ai-dev-atelier
- version
- 1.0
SKILL.md
Git Commit
Write clear, atomic commits. Detect and match project conventions first; use Conventional Commits as fallback. Every commit = one complete, working change.
Feature Branch Workflow
Philosophy: Commit early, commit often.
Feature branch: Create → [Change → Test → COMMIT]* → PR → Merge
Why frequent commits matter:
| Benefit | Description |
|---|---|
| Easy rollback | Revert to last working state in seconds |
| Safety checkpoints | Never lose more than one small step |
| Better debugging | git bisect works with granular history |
| Collaboration | Others see incremental progress |
| Automation | Enables auto-changelogs and semantic versioning |
Anti-pattern: One giant commit at end of feature (47 files, 2800+ lines). Split into logical increments instead.
Quick Start
Pre-Commit Checklist:
- Code builds and tests pass
- ONE logical change (describable in one sentence)
- Checked project conventions (Step 1)
Workflow
Step 1: Detect Conventions (REQUIRED)
ls .commitlintrc* commitlint.config.* .czrc .cz.json 2>/dev/null
git log --oneline -15
| Finding | Action |
|---|---|
| Config file exists | Follow those rules exactly |
feat:, fix: pattern |
Use Conventional Commits |
[JIRA-123] pattern |
Match ticket prefix format |
Component: message pattern |
Match component prefix format |
| No clear pattern | Use Conventional Commits (Step 2) |
Step 2: Format Message
<type>(<scope>)!: <subject> ← 50 chars ideal, 72 max
← blank line
<body> ← wrap at 72 chars, optional
← blank line
<footer> ← optional (Closes #123, BREAKING CHANGE)
Subject rules: Imperative mood (add not added), lowercase after colon,
no trailing period, no filler words.
Types:
| Type | Purpose | SemVer | Example |
|---|---|---|---|
feat |
New feature | MINOR | feat(auth): add OAuth |
fix |
Bug fix | PATCH | fix(cart): prevent neg qty |
docs |
Documentation | - | docs(api): add examples |
style |
Formatting | - | style: fix indentation |
refactor |
Code restructure | - | refactor: extract helper |
perf |
Performance | - | perf(db): add index |
test |
Tests | - | test(auth): add edge cases |
build |
Build/deps | - | build: update webpack |
ci |
CI config | - | ci: add deploy workflow |
chore |
Maintenance | - | chore: update gitignore |
revert |
Revert previous commit | - | revert: let us not... |
Breaking changes: Add ! after type/scope, explain in body/footer.
feat(api)!: require API key for all endpoints
BREAKING CHANGE: Anonymous access removed. All requests need X-API-Key header.
Step 3: Body (When Needed)
| Required | Optional |
|---|---|
| Breaking change (impact/migrate) | Self-explanatory from diff |
| Non-obvious fix (root cause) | Simple docs/style/chore |
| Complex feature (design decision) | Tests with descriptive names |
Body content: WHY (motivation), WHAT problem, HOW users affected. Not: Which files (diff shows), line-by-line explanation (code comments).
Step 4: Footer
Closes #123 # Issue reference
Fixes #456 # Also closes issue
Refs: #789 # Reference without closing
BREAKING CHANGE: <description> # If not in body
Co-authored-by: Name <email@example.com> # Co-authorship
Reviewed-by: Name # Use hyphens in tokens
Commit Granularity
Atomic Commit = Smallest Complete Change
- ONE purpose - single feature/fix/refactor
- Self-contained - doesn't depend on uncommitted work
- Leaves code working - builds pass, tests pass
- Revertable alone - without breaking other features
What Goes Together vs Separate
| Together (same commit) | Separate (different commits) |
|---|---|
| Feature + its unit tests | Feature + unrelated formatting |
| Bug fix + regression test | Bug fix + dependency update |
| API change + docs update | Refactor + new feature |
| Refactor + affected tests | Multiple unrelated fixes |
When to Commit (Triggers)
| Trigger | Action |
|---|---|
| Function works and tested | COMMIT now |
| Test passes (red → green) | COMMIT now |
| Bug fixed and verified | COMMIT before next task |
| About to refactor | COMMIT working state first |
| Starting new sub-task | COMMIT current progress |
| End of session / before pull | COMMIT if working |
Key: Before starting new task, commit all current task changes.
When NOT to Commit
| Situation | Why |
|---|---|
| Code doesn't build | Breaks bisect, blocks others |
| Tests failing* | Not a valid checkpoint |
| "I think this works" (untested) | Verify first, commit second |
| Mid-debugging / mid-fix | Wait until fix is complete |
| Uncertain about approach | Prototype first, commit when solid |
| CI still running / red | Wait for green, then commit fix |
*Exception: TDD RED commits (test-only commits that intentionally fail) are valid.
"Commit often" means verified working increments, not untested guesses.
❌ BAD: Commit-and-pray workflow
───────────────────────────────
1. Write fix
2. "Looks right to me" → commit
3. Push → CI fails
4. "Oops, forgot X" → commit again
5. Push → CI fails again
6. Repeat 3 more times...
Result: 5 broken commits in history
✓ GOOD: Verify-then-commit workflow
───────────────────────────────
1. Write fix
2. Run tests locally → fails
3. Fix the issue
4. Run tests locally → passes
5. Commit
6. Push → CI passes
Result: 1 clean commit
Rule: If CI or local tests are red, you're not done. Fix first, verify, then commit.
TDD Pattern
test(auth): add failing test for validation # RED
feat(auth): implement validation # GREEN
refactor(auth): extract to helper # REFACTOR
Examples
See references/examples.md for comprehensive examples. Quick reference:
# Feature with context
feat(auth): add password strength indicator
Users lacked feedback on password quality during signup.
Add real-time strength meter, requirements checklist, submission block.
Closes #234
# Bug fix with root cause
fix(api): handle null response from payment gateway
Gateway returns null on timeout. Add null check with retry logic.
Fixes #567
# Simple changes (no body needed)
docs(readme): add Docker installation steps
test(utils): add edge case tests for date parser
chore(deps): update lodash to 4.17.21
| Bad | Problem | Good |
|---|---|---|
fix bug |
No context | fix(cart): prevent duplicate items |
update code |
Meaningless | refactor(api): simplify errors |
WIP |
Incomplete | Finish work, then commit |
misc fixes |
Multiple changes | Split into separate commits |
Integration
| When | Skill | Why |
|---|---|---|
| Before committing | code-quality |
Ensure checks pass |
| After committing | docs-check |
Check if docs need update |
| After PR review | pr-review → git-commit |
Resolve then commit |
References
references/examples.md- Extended examples by scenario
Output
Git commits created via standard git commands. No files saved. Commit history visible via git log.
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
LukasStrickler/ai-dev-atelier
resolve-pr-comments
Resolve bot review comments (CodeRabbit, Copilot, Gemini) on GitHub PRs using subagents. Use when: (1) User asks to 'review PR comments' or 'resolve PR comments', (2) User says 'work through PR N comments' or 'handle bot comments', (3) Need to triage CodeRabbit/Copilot/Gemini review comments, (4) Processing PR feedback at scale, (5) Want to see what's already fixed vs still pending. NOT for: creating PRs, reviewing code yourself, writing new reviews. Triggers: review PR comments, resolve PR comments, work through PR comments, handle bot comments, process CodeRabbit comments, triage PR feedback, fix PR review issues, resolve bot comments, pr comment resolver.
LukasStrickler/ai-dev-atelier
tdd
Strict Red-Green-Refactor workflow for robust, self-documenting code. Discovers project test setup via codebase exploration before assuming frameworks. Use when: (1) Implementing new features with test-first approach, (2) Fixing bugs with reproduction tests, (3) Refactoring existing code with test safety net, (4) Adding tests to legacy code, (5) Ensuring code quality before committing, (6) When tests exist but workflow unclear, or (7) When establishing testing practices in a new project. Triggers: test, tdd, red-green-refactor, failing test, test first, test-driven, write tests, add tests, run tests.
LukasStrickler/ai-dev-atelier
use-graphite
Manage stacked PRs with Graphite CLI (gt) instead of git push/gh pr create. Auto-detects Graphite repos and blocks conflicting commands with helpful alternatives. Use when: (1) About to run git push or gh pr create in a Graphite repo, (2) Creating a new branch for a feature, (3) Submitting code for review, (4) Large changes that should be split into reviewable chunks, (5) Hook blocks your git command and suggests gt equivalent. NOT for: repos not initialized with Graphite, git add/commit/status/log. Triggers: git push blocked, gh pr create blocked, create branch, submit PR, stacked PRs, split large PR, gt create, gt submit, graphite workflow.
LukasStrickler/ai-dev-atelier
code-review
Review code changes using CodeRabbit CLI - supports uncommitted files (task mode) or all PR files vs main branch (pr mode). Catches bugs, security issues, and code quality problems before committing or when reviewing pull requests. Use when: (1) Reviewing uncommitted changes before committing (task mode), (2) Reviewing all changed files in a PR against main branch (pr mode), (3) Working on subtasks and want to check progress, (4) Need feedback on work-in-progress code, (5) Preparing PR for merge, (6) When CodeRabbit review is needed, (7) For bug detection and security scanning, or (8) For automated code quality assessment. Triggers: review code, check code quality, review changes, code review, review PR, check for bugs, security scan, review uncommitted, finalize code, pre-commit review.
LukasStrickler/ai-dev-atelier
image-generation
Generate, edit, and upscale AI images. Use when creating visual assets for apps, websites, or documentation. FREE Cloudflare tier for iterate generation (~96/day), Fal.ai for paid tiers. Four quality tiers (iterate/default/premium/max). Supports text specialists, multi-ref editing, SVG, background removal. Triggers: generate image, create image, edit image, upscale, logo, picture of, remove background.
LukasStrickler/ai-dev-atelier
code-quality
Run comprehensive code quality checks including TypeScript typecheck, ESLint linting, Prettier formatting, and Markdown validation. Auto-fixes formatting issues in agent mode or provides read-only checks for CI pipelines. Use when: (1) Before committing code changes, (2) In CI/CD pipelines for automated quality gates, (3) After making significant code changes, (4) When preparing code for review, (5) When ensuring code meets quality standards, (6) For type checking, linting, formatting, and markdown validation, (7) In pre-commit hooks, or (8) For automated quality gates before merging. Triggers: finalize, code quality, typecheck, lint, format, check code, quality check, run checks, pre-commit, before commit, CI checks, validate code.
Didn't find tool you were looking for?