commit-changes
Skill6 starsChanged 50 days agoApache-2.0View on GitHub
仅手动触发。用户明确要求暂存、提交或推送时,先跑项目提交前门禁(若有),审查 diff,逐文件精确 stage 并等用户确认,再按项目提交规范写中文提交信息;禁止 git add -A 和 git add .;未明确要求不推送。触发词:"提交"、"commit"、"push"、"推送"、"帮我提交"。
12 other projects use a version of this.
Adopted by
Most recently taken up by
- vinaySankar2004/vuhritheir own version, first seen Nov 3, 2026
- jimmy-guzman/skillstheir own version, first seen Oct 20, 2026
- gspencergoog/skillstheir own version, first seen Oct 16, 2026
- koruworks/skillstheir own version, first seen Oct 15, 2026
- foldkit/foldkittheir own version, first seen Oct 8, 2026
- Kotlin/kotlinx-rpctheir own version, first seen Oct 8, 2026
Other versions12 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 12
- tldraw/tldraw★ 51ksame name
- foldkit/foldkit★ 1.1ksame name
- Kotlin/kotlinx-rpc★ 1.1ksame name
- dennishenle/agent-skills★ 2same name
- gspencergoog/skills★ 2same name
- browserkit-dev/browserkit★ 1same name
- DejusDevspace/my-skills★ 0same name
- jimmy-guzman/skills★ 0same name
- koruworks/skills★ 0same name
- ltatarev/skills★ 0same name
2 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository26 other files it gives its agents.
- AGENTS.mdAGENTS.md
- adversarial-reviewSkill
- ask-codexSkill
- closed-remediation-reviewSkill
- code-to-7layerSkill
- code-to-guideSkill
- codex-reviewSkill
- controlSkill
- deep-research-gateSkill
- design-previewSkill
- doc-html-styleSkill
- doc-layer-systemSkill
- docs-from-codeSkill
- git-worktreeSkill
- goal-charterSkill
- legacy-archaeologySkill
- lightweight-designSkill
- long-doc-governanceSkill
- one-by-oneSkill
- plan-goalSkill
- read-togetherSkill
- site-previewSkill
- task-control-docSkill
- test-case-designSkill
- test-execution-routerSkill
- test-standardsSkill
The file
Files
Commit Changes
用户明确要求提交、推送或说"帮我提交"时使用。本 skill 是用户级通用引擎:只固定流程与红线;提交规范原文、门禁脚本、额外前缀规则由项目注入,本 skill 不硬编码任何项目值。
项目注入(挂载点)
| 槽位 | 来源(按顺序探测,先命中先用) | 缺省 |
| 提交规范原文 | 项目指令文件(AGENTS.md)的「提交规范」节 → CONTRIBUTING.md | 本 skill「Rules」的默认格式 |
| 提交前门禁命令 | 项目补丁 skill 显式声明的命令 → .agents/hooks/pre-handoff-check.sh → .agents/hooks/pre-commit-check.sh | 无门禁,跳过 Step 0 |
| 额外首行规则 | 提交规范里声明的特殊前缀(如跨域提交标记、任务编号) | 无 |
Required reads
- 上表命中的提交规范原文——每次都读,不凭记忆写格式。
Rules
- 提交信息用简体中文,格式
<类型>: <简要描述>;项目提交规范另有规定时以项目为准 - 提交前必须读 diff
- 未经用户明确要求,不 amend、不 force-push、不 push
- 不提交自己不理解、与本次任务无关的改动
- 永远不用
git add -A/git add .
Default flow
Step 0 — 门禁(项目有门禁脚本时必跑,不可跳过)
# 按挂载点顺序探测;命中哪条跑哪条,全部不存在则跳过本步
bash .agents/hooks/pre-handoff-check.sh- 退出码非 0 → 立即停止,告知用户失败项,等修复后再触发本 skill
[WARN]输出 → 记录警告,在 Report back 中提及,不阻断流程
Step 1 — 确认变更文件清单
git status --short读取输出,理解当前工作区状态(哪些是 tracked 修改、哪些是 untracked 新文件)。
Step 2 — 阅读 diff,理解改动性质
git diff
git diff --cached判断:改动主题(功能/修复/文档/重构/规范/配置)、涉及模块、是否含无关改动需拆分。
Step 3 — 拟定 stage 清单并等待确认
[!CAUTION] 禁止
git add -A或git add .。必须逐文件或逐目录精确 stage。
列出拟 stage 的文件清单(git add <具体路径> 形式),明确告知用户并等待确认,用户确认后再执行 stage。
Step 4 — 拟写 commit message
默认格式:<类型>: <简要描述>,类型限:修复 / 功能 / 重构 / 规范 / 文档 / 配置。
项目提交规范声明了额外首行规则时优先遵守(例:一次提交改动多个业务域目录时,首行用项目规定的跨域标记并说明原因)。
Step 5 — 执行提交
git commit -m "<commit message>"Step 6 — 推送(仅在用户明确要求时执行)
用户未说"push"或"推送"时,不自动推送。
Report back
用中文总结:
- staged 了哪些文件(列出完整路径)
- commit message 是什么
- 是否已 push
- 若门禁脚本有
[WARN]输出,说明具体警告内容 - 若接口面有变更但未更新测试脚本,说明原因
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

