source/skills/git-commit/SKILL.md
Git commit workflow that analyzes recent history to match repo-specific message conventions. Use when you don't have recent commit history in context and need to determine the repository's commit message style, or when crafting commit messages that need to match existing conventions.
npx skillsauth add scruffydan/AI-Agents git-commitInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
This skill ensures commit messages follow the repository's established conventions by analyzing recent commit history.
Run this when you don't have recent commit history in context
git log -10 --oneline --no-decorate
Analyze for patterns:
feat(api): vs feat:)For deeper pattern analysis:
git log -10 --format="%h %s%n%b%n"
Look for:
Use the EXACT same conventions as recent commits:
<type>(<scope>): <description>
If commits are free-form, use the same free-form style without prefixes.
Follow repository conventions discovered in steps 1-3:
git commit -m "type(scope): description"
# Or with body
git commit -m "type(scope): description" -m "
Detailed explanation of changes and reasoning.
Closes #123"
Focus on WHY, not WHAT:
Be specific and actionable:
Keep subject line concise:
Use imperative mood (if repository does):
When changes are complex:
git commit -m "feat: add payment processing" -m "
Explain why, impact, and any references."
git status
git diff --cached
Ensure:
# See what will be committed
git diff --cached --stat
# Verify commit message locally
git commit --dry-run
Before committing:
After committing:
git-push skill for pre-push checklistBefore running git commit:
testing
Use when creating new skills, editing existing skills, or verifying skills work before deployment
data-ai
Use when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidence before assertions always
development
How to use @docs-fetcher to pull current, targeted external documentation (APIs, SDKs, configs, errors, version changes). Use when you need authoritative docs or examples instead of relying on memory.
development
How and when to invoke code review agents (@code-security, @code-readability, @code-performance, @code-redundancy, @code-simplifier, @code-full-review). Use before major PRs, audits, refactors, or whenever targeted review feedback is needed.