Analyzes git changes and creates conventional commit messages with branch suffixes.
npx skills add https://github.com/Cohey0727/CodingAgentTools --skill commit-cohey0727
Please help me install this Agent Skill. Skill: commit Source: https://github.com/Cohey0727/CodingAgentTools/tree/main/skills/commit Command: npx skills add https://github.com/Cohey0727/CodingAgentTools --skill commit-cohey0727
<type>: <comment> @<branch>, written in the repository's dominant language. - Safety Guardrails: Never stages secrets, never uses git add -A without confirmation, never amends or pushes without explicit approval, and runs lint/tests before committing when configured. - Use Case: After finishing a bug fix, simply ask to commit your changes and receive a properly formatted message like fix: 決済プロバイダのnullレスポンスを処理 @fix/payment-webhook without writing it yourself. ## Quick Start Ask the assistant to commit your current changes, and it will analyze the diff, propose a conventional commit message, and execute the commit after your confirmation.High-intent search queries and answers about installing and using this skill.
Stage your changes and invoke the commit skill. It analyzes the git diff, classifies the change as feat, fix, refactor, or chore, and generates a subject-only message in the format `<type>: <comment> @<branch>` for your approval before committing.
The skill never runs `git add -A` or `git add .` without confirmation. If nothing is staged, it asks which specific files to stage by name and refuses to stage files that likely contain secrets such as `.env` or credentials files.
Yes, if the project defines lint or test scripts in files like package.json or a Makefile, the skill runs them before committing. If they fail, it fixes the issues first; projects without such configuration skip this step.
It can, but only with explicit user confirmation. After committing, it checks for an upstream branch and unpushed commits, then asks whether to push or set an upstream with `git push -u origin <branch>`.
Messages follow the repository's dominant language, detected from existing commit history and documentation. Japanese repositories get Japanese comments, English repositories get English, and an explicit user language request overrides detection.
The skill never bypasses hooks with `--no-verify`. When a hook fails, it fixes the underlying issue and creates a new commit rather than amending the failed one.
NUMU-IOReviews git staged code changes for security vulnerabilities and generates prioritized ODAWP action plans.
VYRE-StudiosCommit staged changes, push the branch, and open a pull request via GitHub CLI.
john-data-chenAutomate pre-commit code verification with static security scans, linting, and reviewer subagent workflows.
laicluseGroups all uncommitted git changes into logical commits with descriptive messages.