AgentSkillsCN

git-commit-messages

按照项目规范生成一致的 Git 提交信息。当用户提交代码变更、创建拉取请求,或被要求撰写提交信息时,可使用此技能。

SKILL.md
--- frontmatter
name: git-commit-messages
description: Generates consistent git commit messages following project conventions. Use when committing changes, creating PRs, or when asked to write commit messages.

Git Commit Message Generator

Format

code
type(scope): subject in imperative mood

- Body bullet in past tense with period.
- Another change description.

Types

TypeWhen to Use
featAdded new functionality
fixFixed a bug
refactorRestructured code, no behavior change
choreDependencies, tooling, configs
docsDocumentation
testTests
cicdCI/CD pipelines, deployment, .github/ configs (workflows, dependabot, etc.)
aiAI/Claude configurations

Rules

  1. Subject: Imperative mood, lowercase after colon, no period, max 72 chars
  2. Scope: Derived from path (e.g., apps-server, apps-expo, shared-domain, scripts, services). When changes span multiple scopes, omit the scope entirely
  3. Body: Past tense, capital start, period at end
  4. No attribution: Never include "Co-Authored-By", "Generated with", or any AI/author attribution
  5. AI-only changes: When changes are exclusively AI-related (skills, prompts, Claude configs), always use ai type—never refactor, chore, or other types
  6. Preview before commit: Always show the proposed commit message to the user for confirmation before executing the commit

Examples

code
feat(apps-server): add health check endpoint
code
refactor(apps-server): migrate from LibSQL to PostgreSQL

- Replaced LibSQL/Turso with PostgreSQL for Better Auth storage.
- Removed OpenFGA environment variables.
- Added pg and Effect SQL dependencies.
code
chore: update workspaces and dependencies

- Added shared/repositories/* to workspaces.
- Bumped @biomejs/biome to 2.3.8.
code
fix(apps-cli): update build filter to exclude xiroi-apps directory

- Changed the build command filter from excluding './apps/*' to excluding
  './xiroi-apps/*' for more accurate targeting.
code
ai: secure Claude settings by restricting dangerous permissions