无需联网,准备提交、起草提交信息与变更摘要
Scanned 9/2/2026
Install to Claude Code
npx -y skills add wangzhe04/ruyi-workbench-oss --skill commit-workflow --agent claude-codeInstalls into .claude/skills of the current project.
Are you the author of Commit Workflow?
Add the live security badge to your README — it updates automatically with every re-scan.
[](https://www.skillsdirectory.com/skills/wangzhe04-commit-workflow)More formats (shields.io, HTML) on the badges page.
---
name: 提交工作流
description: 无需联网,准备提交、起草提交信息与变更摘要
---
# Commit Workflow
Use this skill for local commit preparation, commit message drafting, and changelog summaries without GitHub or network access.
Workflow:
1. Run `git_status` for the overview, then `git_diff` for the change content. (`git_commit` creates the commit — step 6.)
2. Group changes by user-facing behavior, tests, docs, and infrastructure.
3. Read important changed files before summarizing risk.
4. Run the local test command if one is obvious from `dependency_inventory`.
5. Draft a concise commit message in the repository's style.
6. Only create the commit when the user explicitly asks for it.
Message format:
- Use an imperative subject line under 72 characters.
- Add a short body when the change spans multiple areas.
- Mention tests run and important caveats outside the commit message if not committing.
Do not:
- Push, fetch, or open a remote.
- Stage unrelated user changes.
- Hide generated files, lockfile changes, or skipped tests.
Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.
No comments yet. Be the first to comment!
Adversarial multi-agent planning skill. Self-orchestrates 5 hostile category members (unspecified-low, unspecified-high, deep, ultrabrain, artistry) via team-mode for ruthless cross-critique debate, distills only the defensible insights, then MANDATORILY hands the distilled insight bundle to the `plan` agent for executable plan formalization. Use when planning needs maximum rigor and surfacing of weak assumptions, blind spots, and over-engineering. Triggers: 'hyperplan', 'hpp', '/hyperplan', ...