Token导航 LogoToken导航TokenDH.com
开发需要联网github未标认证来源可访问许可证需确认审计通过

semantic-commit语义提交

Agent Skill

semantic-commit 用于处理 GitHub 仓库、Issue、Pull Request 和代码协作信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要围绕仓库状态、代码变更或协作事项进行整理时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

659

周安装

28

GitHub Stars

1

下载量

231
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

unknown

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:semantic-commit(语义提交)
来源仓库:https://github.com/buildrtech/dotagents
仓库路径:skills/semantic-commit
安装命令:
npx skills add https://github.com/buildrtech/dotagents --skill semantic-commit
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/buildrtech/dotagents --skill semantic-commit

简介

用于处理 GitHub 仓库、Issue、Pull Request 和代码协作信息。

  • 适合在需要围绕仓库状态、代码变更或协作事项进行整理时使用。
  • 可结合来源仓库和原始 README 核验具体用法。
  • 安装方式:通过 npx skills add 命令从指定 GitHub 仓库添加。
  • 注意权限范围和维护状态,确认是否会触发联网、命令执行或文件读写。

SKILL.md

Semantic Commit

Create git commits following the Conventional Commits specification.

Commit Format

<type>[optional scope]: <description>

[optional body]

[optional footer(s)]

Guidelines

  • Start with a short prefix followed by colon and space (feat:, fix:, docs:, refactor:, test:, chore:, etc.)
  • Isolate your changes and commit only what you did (no unrelated files)
  • feat: for user-visible features, fix: for bug fixes
  • A scope MAY be added in parentheses, e.g. fix(parser): - only when it meaningfully improves clarity
  • Short description in imperative mood explaining what changed, not how
  • Body MAY be included after one blank line for context, rationale, or non-obvious behavior
  • Footers MAY be included (Token: value format, use - instead of spaces in tokens)
  • Breaking changes should be explained clearly in description or body, no special marking required
  • Clarity and usefulness matter more than strict conformance

Commit Types

TypeDescription
featA new feature (user-visible)
fixA bug fix (user-visible)
docsDocumentation only changes
styleChanges that do not affect the meaning of the code (formatting, semicolons, etc.)
refactorA code change that neither fixes a bug nor adds a feature
perfA code change that improves performance
testAdding missing tests or correcting existing tests
buildChanges that affect the build system or external dependencies
ciChanges to CI configuration files and scripts
choreOther changes that don't modify src or test files
revertReverts a previous commit

Workflow

digraph commit_workflow {
    rankdir=TB;
    node [shape=box];

    check [label="Check git status"];
    stage [label="Stage changes\n(if needed)"];
    review [label="Review staged diff"];
    analyze [label="Analyze changes\nDetermine type & scope"];
    write [label="Write commit message"];
    commit [label="Execute git commit"];

    check -> stage;
    stage -> review;
    review -> analyze;
    analyze -> write;
    write -> commit;
}

1. Check Repository State

git status --short
git diff --cached --stat   # staged changes
git diff --stat            # unstaged changes

2. Stage Changes

If no changes are staged:

  • Ask which files to stage
  • Stage only files you changed for this task (avoid unrelated changes)

3. Review Staged Changes

git diff --cached

Understand what was modified to write an accurate message.

4. Determine Type and Scope

Based on the changes:

  • Type: What category of change? (feat, fix, refactor, etc.)
  • Scope: What area of the codebase? Only add when it meaningfully improves clarity
  • Breaking: Does this break backward compatibility? Explain clearly in description or body

Writing Good Commit Messages

Subject Line

  • Use imperative mood: "add feature" not "added feature"
  • Keep under 50 characters when possible
  • Don't end with a period
  • Explain what changed, not how

Body (Optional)

Include after one blank line when context, rationale, or non-obvious behavior needs explanation:

  • Explain motivation for the change
  • Contrast with previous behavior
  • Note any side effects
fix(parser): handle unicode characters in filenames

Previously, filenames with non-ASCII characters would cause
a decode error. Now using UTF-8 decoding with fallback to
latin-1 for legacy files.

Closes #123

Footer (Optional)

Use Token: value format (use - instead of spaces in token names):

  • Closes #123 or Fixes #456 for issue references
  • Co-authored-by: Name <email> for pair programming
  • Reviewed-by: Name <email> for review attribution

Examples

ChangesCommit
New endpoint addedfeat(api): add user profile endpoint
Bug causing crashfix: prevent null pointer on empty config
Updated READMEdocs: add installation instructions
Reformatted codestyle: apply prettier formatting
Renamed internal functionrefactor: rename processData to parseInput
Optimized queryperf(db): add index for user lookups
Added unit teststest: add coverage for auth module
Updated dependenciesbuild: upgrade react to v18
Changed CI configci: add node 20 to test matrix
Cleaned up fileschore: remove unused imports
Breaking API changefeat(api): change response format to JSON arrays

Common Mistakes

MistakeFix
Past tense ("added")Use imperative ("add")
Too vague ("fix bug")Be specific ("fix null check in parser")
Too long subjectKeep concise, move details to body
Wrong typefix = bug, feat = new capability, refactor = no behavior change
Combining unrelated changesSplit into multiple commits
Overusing scopeOnly add scope when it meaningfully improves clarity

Quick Reference

feat:     New feature for users
fix:      Bug fix for users
docs:     Documentation only
style:    Formatting, no logic change
refactor: Code change, no behavior change
perf:     Performance improvement
test:     Adding/fixing tests
build:    Build system, dependencies
ci:       CI configuration
chore:    Maintenance, tooling
revert:   Reverting commits

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Codex

34.36%
按下载量换算79

Claude

27.38%
按下载量换算63

Cursor

20.4%
按下载量换算47

Gemini CLI

9.58%
按下载量换算22

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills