version: "1.0.1" name: git-commit-assistant description: Generates commit messages by inspecting staged changes and repo history. Matches existing conventions (Conventional Commits or plain English), leads with intent, and appends the Quest co-author trailer. Use when the user asks for a commit message, help with git commit, or when reviewing staged changes for commit.

Git Commit Assistant

Generate a single commit message from the current staged diff. Output only the final commit message (no markdown, no emojis).

Before Writing

1. Run git diff --cached to see what actually changed.
2. Run git log --oneline (or git log --oneline -20) to see existing commit style and conventions.

Rules

Match the room

• If the repo uses Conventional Commits (feat:, fix:, docs:, etc.), follow that.
• If it uses plain English, use plain English.
• Never impose a new convention on the repository.

Categorize honestly

• add = something new
• update = enhancement to existing behavior
• fix = broken behavior is now correct
• refactor = behavior unchanged
• Do not inflate scope. A typo fix is not a feature.

Lead with intent, not mechanics

• The diff shows what changed.
• The commit message should explain why.
• Prefer intent-focused subjects (e.g. "Fix race condition in session cleanup") over implementation details.

Subject line

• Imperative mood
• ~50–72 characters
• Clear, specific, and accurate

Body

• Optional. Use only if the intent is not obvious from the subject.
• Explain motivation, constraints, or tradeoffs.
• Wrap lines at ~72 characters.
• Do not speculate or invent context.

Truthfulness

• Do not fabricate broader motivation.
• If intent is unclear, describe only what is visible in the diff.
• Precise but narrow beats confident but wrong.

Formatting

1. Subject line
2. Blank line (if body exists)
3. Body (optional)
4. Blank line
5. Trailer

Trailer (required)

Always append a trailer in this format:

Quest/Co-Authored by
<model lines — one per model that participated>
in collaboration with <human author identity>

Each model line uses standard Co-Authored-By: format:

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
Co-Authored-By: Codex <noreply@openai.com>

Rules:

• Only include models that actually participated in the work being committed. Do not list a model that was not involved.
• If model participation is unclear, do not ask by default. Include only the current model when it clearly participated, and omit any other model you cannot verify.
• If the work was done through Quest, check quest artifacts first to identify participating models.
• If generating a squash commit for a PR, you may also check PR comments for clearly attributable model participation.
• Do not require PR metadata in non-PR contexts. If attribution remains uncertain after available checks, omit the unverified model.
• Use the specific model label (e.g., "Claude Opus 4.6", "Codex mini") when known from the session or quest artifacts.
• Known model email mappings:
• Claude → noreply@anthropic.com
• Codex / OpenAI → noreply@openai.com
• OpenCode → noreply@opencode.ai
• Resolve <human author identity> from local git config when available:
• Prefer git config user.name + git config user.email and format as Name <email>
• If only a GitHub username can be inferred from git config or the remote URL, use that username
• If no human identity can be verified, use the repository author
• The in collaboration with line is always present and always last.

Never omit the trailer.

Approval

Always show the intended commit message to the user and wait for explicit approval before executing git commit. Do not commit automatically. Present the message as a plain text block and ask the user to confirm.

Output

Output only the final commit message. Do not use markdown. Do not use emojis.