git-commit

name	git-commit
description	Safely orchestrate Conventional Commits for staged Git changes, or all working-tree changes when the user explicitly asks to include everything. Use when asked to write a commit message, split staged or working-tree changes, organize a messy index, or draft commit text without pushing. Auto-detects commit language; explicit phrases like 用中文提交 or 请使用中文拆分提交所有的改动 override. Default headers have no [AI] tag; add it only when asked (加 AI 标记 / add [AI] tag / 不要 [AI]). Create commits with local git commit only — never GitHub web, GitHub file-commit APIs, or a platform GitHub App — so GitHub does not show "This commit was created on GitHub.com" or "Committed via Cursor Agent". 本地 git 提交; 不要在 GitHub 上创建 commit; 不要 Committed via. Not for push, pull-request creation, amend, rebase, or tag operations.
category	git-github-collaboration
tags	["git","conventional-commits","commit-message","agent-aware"]
version	1.12.0
allowed-tools	Read, Bash

Use this workflow in order: preflight -> split plan -> classify -> compose -> commit/draft -> verify.

Decide the active change authority and output language before doing anything else:

• staged-only is the default. Respect the current index and treat unstaged or untracked files as context only.
• all-changes is allowed only when the user explicitly asks to include everything, such as "all changes", "所有改动", "全部改动", "不管有没有 stage", or "包括未跟踪文件". In this mode, the skill may rebuild the index from the full working tree and should treat any existing partial staging as intentionally overridden by the user.
• commit-language is detected, not fixed-default, because forcing one language fights repos (and users) whose history is already in the other. Resolve it in this order, then use it for scope, subject, body, and explanatory output:
  1. An explicit instruction in the user's message — "use English", "英文提交", "用中文", "中文提交", or the legacy phrase 请使用中文拆分提交所有的改动. Honor whichever language the user names.
  2. The dominant language of the user's current request message.
  3. The repository's own habit, sampled in Preflight from git log -n 20 --format=%s. Match what the repo already does so your commit doesn't read as an outlier.
  4. Fall back to English only when none of the above gives a clear signal. Language is orthogonal to emoji, [AI], and trailers: detecting Chinese never changes whether [AI] or agent trailers attach.
• agent-mode is on by default whenever this skill runs (the caller is an agent). It attaches Agent-Task / Agent-Model / Generated-By trailers and applies the Why-line rule for feat / fix / refactor / perf. It does not inject [AI] in the header. Pass --ai only when the user explicitly says "add [AI] tag", "加 AI 标记", "加上 [AI]", or equivalent. Turn agent-mode off when the user says "no AI tag", "不要 AI 标记", "不加 agent trailer", or equivalent.
• commit-channel is local git only. Create the commit with git commit -F <message-file> (or rtk git commit -F). Do not create commits through GitHub's web UI, Contents API, Git Data API, GraphQL createCommitOnBranch, gh api file-commit helpers, GitHub MCP create_or_update_file / push_files, or a platform GitHub App (Cursor Agent, Copilot, Grok[bot], Devin). Those channels make GitHub show "This commit was created on GitHub.com." and often "Committed via ". Do not pass for , , or "Committed via" text.

1. Preflight

1. Inspect git status --short first. Then inspect the active change set:
   • staged-only: git diff --staged --stat and git diff --staged
   • all-changes: git diff --stat, git diff, git diff --staged --stat, and git diff --staged If rtk is available, prefer rtk git status, rtk git diff --staged, and rtk git diff for model-visible inspection.
2. Explicitly note:
   • the active change authority (staged-only or all-changes)
   • staged changes ready to commit
   • unstaged changes that might make the index misleading
   • untracked files that are intentionally excluded vs accidentally forgotten
3. Branch immediately on preflight results:
   • staged-only + no staged changes: stop and tell the user to stage files first.
   • all-changes + no staged, unstaged, or untracked changes: stop and say there is nothing to commit.
   • The active change set exists but is obviously mixed and cannot be safely separated from inspection alone: do not improvise a commit. Output a split plan and stop.
   • The user explicitly asked only for commit text, a draft, or suggestions: continue through classification and composition, but do not run git commit.
4. Detect agent context (skip when user disabled agent-mode):
   • Resolve agent-model from the id of the model currently running this skill — read the runtime's own id (e.g. claude-fable-5 when that is the running model), never a model id copied from an example, which goes stale. This value is required.
   • Resolve agent-task by trying, in order: (a) explicit task ID or issue URL in the user message, (b) closes #N / refs #N mentioned by the user, (c) ticket ID extracted from the current branch name, (d) Agent-Task value from the previous commit on this branch, (e) fallback to unspecified.
   • Resolve agent-prompt-ref only when a stable prompt reference exists; otherwise leave empty.
   • Detect checkpoint-mode: triggered by user words such as , , , , , , .

2. Split Plan

1. Treat split planning as a safety step, not a formatting nicety.
2. Group the active change set by one coherent unit at a time, usually one of:
   • a single feature or fix
   • one module or package
   • code vs docs vs tests when staged together by accident
3. In all-changes mode, it is acceptable to ignore the current staged subset only because the user explicitly asked to include everything.
4. Atomic check (skip in checkpoint-mode). For every candidate commit, answer:
   • Can the repo compile / tests pass at this commit?
   • Can git revert <sha> undo it without leaving the repo inconsistent?
   • Can one subject line + one Why line explain its intent? If any answer is no, return to the split-plan layer instead of committing.
5. If the split is clean and obvious, explain the planned commit boundaries before composing messages. In all-changes mode, say whether execution will rebuild the index per commit.
6. If the split is ambiguous, stop at the plan. Do not perform hunk-level surgery, edit files, or guess hidden intent just to manufacture atomic commits.
7. Checkpoint mode branch: when checkpoint-mode is detected, skip the atomic check and prepare a single chore(wip): 🔧 [WIP] <subject> commit covering the active change set. Skip Why enforcement. Still attach agent trailers. Do not add [AI] unless the user opted in. In §6 Verify, remind the user to squash [WIP] commits before merging.
8. Read references/split-strategy.md when deciding whether the active change set is safe to keep together.

3. Classify

1. Choose type, optional scope, emoji policy, output language, [AI] policy (off unless the user opted in), Why policy, and whether ! / BREAKING CHANGE is required.
2. Use references/commit-types.md for type and emoji mapping.
3. Use references/message-rules.md for subject, body, footer, issue, breaking-change, and Why rules.
4. Use references/agent-workflow.md for agent context resolution, trailer ordering, checkpoint handling, the local-git commit channel, and the repo-convention branch: when Preflight §5 found an existing AI attribution convention (Assisted-by: / Generated-by:), emit the repo's trailer via --footer-line and omit this skill's private Agent-* / Generated-By: agent group. Never emit host/client attribution (Co-authored-by: Cursor, Made-with, "Committed via").
5. Default to emoji because this repository expects it. Opt out when the user requests no emoji, or when Preflight §5 found a target repo whose history carries no gitmoji.
6. Infer scope from the changed paths instead of inventing one: take the common parent of the changed files. In this repo's layout skills/<category>/<name>/… → scope <name>, and platforms/<platform>/… → scope <platform>. Changes spanning unrelated top-level areas are a signal to split (§2), not a reason to pick a vague umbrella scope. Prefer a scope that already appears in git log when it fits.
7. Keep type in English — the Conventional Commit keyword is the one machine-parsed token and stays stable across languages.
8. Render scope, subject, body, and explanatory output in the commit-language resolved in §0 (user instruction → request language → repo history → English fallback).
9. Mark Why-required when type ∈ {feat, fix, refactor, perf} and checkpoint-mode is off.

4. Compose

1. Resolve the helper path once instead of hardcoding an interpreter name. <skill-dir> below means this skill's base directory — the path announced when the skill loads. It is not an environment variable, so substitute the literal path:
   • Bash / zsh / macOS / Linux: COMMIT_COMPOSER="<skill-dir>/scripts/compose_commit_message"
   • PowerShell: $COMMIT_COMPOSER = "<skill-dir>/scripts/compose_commit_message.ps1"
2. Generate the final message with the wrapper script:
   • Bash / zsh / macOS / Linux: bash "$COMMIT_COMPOSER" ...
   • PowerShell: & "$COMMIT_COMPOSER" ... The wrapper auto-detects python3, python, or py, so do not write python ... directly in the compose step.
3. Required arguments:
   • --type for the commit type; choose a key from references/commit-types.md, or pass the repo's own custom type (lowercase ^[a-z][a-z0-9-]*$, e.g. hotfix) when Preflight §5 found a repo config that defines it. Custom types have no built-in emoji — pair them with --emoji or accept the no-emoji output.
   • --summary for the subject line content. The flag is --summary, not --subject; passing --subject is not accepted and fails argparse.
4. Optional arguments:
   • --body-line for body content
   • --why for the motivation line (rendered as Why: <text> at the top of body)
   • --closes for closing issues
   • --refs for non-closing issue references
   • --footer-line for other structured trailers such as Jira references or a repo's own AI attribution trailer (§3)
   • --confidence / --scope-risk / --tested for the quality-trail trailers (Confidence: / Scope-risk: / Tested:)
   • --breaking-header when the header itself must include !
   • --breaking when a BREAKING CHANGE: trailer is needed

5. Commit Or Draft

1. If the user asked only for a draft, return the proposed commit text and stop.
2. Execution consent checkpoint: Before any git commit, display the final commit message (header + body + footer) and the list of files to be committed. Explicitly call out that the channel is local git commit -F (not GitHub web/API), whether [AI] is in the header (default: no), whether Why is present, and which agent trailers will attach. Then decide whether to pause:
   • Proceed without an extra confirmation when the user's current request already authorizes execution, such as "commit it", "execute the commit", "commit all changes", "直接提交", "提交了", "按方案执行", or 请使用中文拆分提交所有的改动, and preflight found no secret/large-binary risk, no ambiguous split, no missing Why, and no draft-only wording.
   • Wait for explicit confirmation when execution was not clearly requested, when the user asked to review/plan/draft first, when the proposed file set or message differs materially from the requested scope, or when any safety gate in this workflow says to ask before committing.
3. If the user asked to commit and staged-only is active, commit only the safe staged set. Write the message file with the composer's --output flag — e.g. --output "$(git rev-parse --git-dir)/COMMIT_MSG_SKILL" — then commit with git commit -F <message-file> so PowerShell and POSIX shells behave consistently. Invoke the git binary directly. Do not add --trailer. Do not create the commit by editing files through GitHub's Contents API or MCP push_files / create_or_update_file. Never capture the composer's stdout with > redirection in PowerShell: Windows PowerShell 5.1 writes UTF-16LE by default, and git commit -F then reads a Chinese or emoji message as mojibake; --output always writes UTF-8. Keep the message file outside the working tree — the git dir as above or the OS temp dir — never inside the repo, where the all-changes flow's git add -A would sweep it into the commit.
4. If the safety scan found large/binary files, wait for an explicit decision for each risky path before any broad staging command. If the correct outcome is to ignore generated artifacts, edit or ask for a .gitignore update first, then re-run preflight so ignored files are no longer part of the candidate set. Include .gitignore only when the user approved that ignore policy.
5. If the user asked to commit and all-changes is active for a single atomic commit, run only after the safety scan is clear or all risky files have explicit include/ignore decisions, so tracked, deleted, and safe untracked non-ignored files enter the commit set.

6. Verify

1. Read the git commit output before claiming success.
2. Distinguish two hook outcomes before reacting:
   • Hook rejected the commit (non-zero exit, message-format or lint failure): stop and report the original hook failure. Do not silently rewrite the message unless the output clearly says the format is invalid and the user asked you to fix it.
   • Hook rewrote files (a formatter such as prettier/black/gofmt modified tracked files and left them unstaged, aborting or staling the commit): re-inspect git status, re-stage the hook's edits, and retry the same commit message. Say that the hook reformatted files — do not treat the reformatting as your own change.
3. After a successful commit, inspect git log -1 --format='%an <%ae>%n%cn <%ce>%n%B' before claiming success. If the message contains Co-authored-by, Made-with, or Committed via, or the committer is GitHub <[e-mail]>, the wrong channel or a host wrapper added provenance GitHub will render as "This commit was created on GitHub.com." / "Committed via …". Report that outcome. Do not amend (amend is out of scope). Then summarize:
   • the final header
   • whether staged-only or all-changes mode was used
   • the resolved commit-language and which signal chose it (user instruction / request language / repo history)
   • whether emoji was included
   • that [AI] is absent unless the user opted in, and which Agent-* trailers attached
   • which quality trailers attached (Confidence / Scope-risk / Tested), if any
   • whether the commit is a chore(wip) checkpoint
   • whether a Why line is present (required for feat/fix/refactor/perf)
   • whether untracked files were included
   • whether issues or breaking changes were attached
   • that the commit was created with local git commit -F, not GitHub web/API
4. If you stopped before committing, say exactly why: no active changes, no staged changes under staged-only, ambiguous split, Why missing for Why-required type, or draft-only request.
5. If the branch now contains multiple chore(wip): commits, remind the user to squash them via git rebase -i <base-branch> before merging — but do not run rebase from this skill.

References

• references/commit-types.md for commit type, emoji mapping, and end-to-end agent commit examples
• references/message-rules.md for message structure, Why-line rule, trailers, and agent trailer field table
• references/split-strategy.md for split heuristics, atomic-check three questions, and checkpoint vs atomic distinction
• references/agent-workflow.md for agent context resolution, local-git commit channel, checkpoint mode, and audit commands