- Home
- Skills
- Jiatastic
- Open Python Skills
- Commit Message
commit-message_skill
- Python
2
GitHub Stars
1
Bundled Files
2 months ago
Catalog Refreshed
4 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill jiatastic/open-python-skills --skill commit-message- SKILL.md5.0 KB
Overview
This skill analyzes git changes and generates Conventional Commit-style messages tailored to the actual code edits. It supports generating single commit messages and splitting large unrelated changes into batch commit suggestions. Use it to produce concise, scope-aware commit headlines and useful bullet-point bodies that reflect business intent.
How this skill works
The tool inspects changed files (or staged changes) and groups them by directory, change type, and semantic relationship to propose logical commit groups. For each group it suggests a Conventional Commit type, scope, short subject, and a short body with key points. Use the analyzer output as a starting point, then read diffs to refine messages so they reflect the real code and business context.
When to use it
- Creating a single commit from staged or unstaged changes
- Reviewing staged changes before committing to ensure clear intent
- Splitting a large changeset into multiple logical commits
- Enforcing Conventional Commits in a collaborative repo
- Generating commit messages for CI or automation when diffs are available
Best practices
- Read the actual git diff before finalizing the suggested message to capture intent
- Choose scope as a business module or feature (auth, api, calendar), not just a folder name
- Keep the subject under 72 characters and use active verbs (add, fix, update)
- Group related file edits together; use --batch to surface unrelated groups
- Include short bullet points describing the important functional changes in the body
Example use cases
- Add a new API endpoint: generate feat(api): add user creation endpoint with validation details
- Fix a token refresh bug: produce fix(outlook): refresh OAuth token before sending emails
- Split a monolithic changeset into focused commits using --batch
- Document dependency bumps: chore(deps): update requests and pytest versions
- Refactor internal helpers across modules: refactor(utils): extract rate limiting middleware
FAQ
By default it analyzes all changes, and you can pass --staged to limit analysis to staged files.
How does batch mode work?
Batch mode groups unrelated changes by directory, change type, and semantic relationship, then suggests separate commit messages for each group.