git-commit-assistant
Git Commit Assistant
Quick start
- Review staged changes and ensure they cover a single concern.
- Choose
<type>(optional-scope)and write an imperative 50-72 character summary. - Add body for why/decisions/trade-offs; footers for breaking changes or issue references.
- Show the message in a fenced code block, ask for approval to commit, then commit only if approved.
- See
references/commit_rules.mdfor full rules and examples.
Workflow
-
Inspect changes
- Check
git statusandgit diff --cachedto confirm what will be committed. - If changes mix concerns or include formatting noise, ask to split or stage appropriately.
- Check
-
Select type/scope
- Allowed types: feat, fix, refactor, docs, test, chore, build, ci, perf, style.
- Scope is optional; keep it short (e.g.,
api,auth,deps).
-
Draft the message
- Summary: imperative, no trailing period, describe what/why, not implementation detail.
- Body (only if valuable): why the change was needed, key decisions, trade-offs; wrap to ~72 chars.
- Footers (when relevant): breaking changes, migration notes, issue references (e.g.,
Closes #123).
-
Quality gate
- Message intent must match the staged diff; avoid kitchen-sink commits.
- Do not mix formatting-only changes with functional work.
- Avoid noisy summaries like
update,fix stuff,changes,wip.
Confirmation and commit
- Present the message in a fenced code block and ask the user to accept.
- On approval:
- Confirm staged changes are correct; surface unstaged items before committing.
- Run
git commit -m "<summary>"with additional-m "<body>"/-m "<footers>"as needed. - If the commit fails, report the error and pause for guidance.
- If the user declines, revise the message or wait for more information.
Reference
references/commit_rules.md: conventional format, output pattern, quality gate, example output.
More from dmitriiweb/extract-emails
google-docstring-assistant
Write Python docstrings following the Google Python Style Guide, using clear sections and examples.
10format-lint-assistant
Run the project's formatter, linters, and mypy checks in the required order, fixing issues and managing any needed stub dependencies via uv.
4skill-creator
Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Claude's capabilities with specialized knowledge, workflows, or tool integrations.
3pytest-testing-assistant
Write focused pytest tests as standalone functions (one test per function), avoiding test classes.
3code-review-assistant
Perform structured, actionable code reviews for Python code with clear findings and suggestions.
3debug-logging-assistant
Add purposeful debug logging to improve observability without changing behavior.
3