Token导航 LogoToken导航TokenDH.com
待分类只读github未标认证来源可访问许可证需确认审计通过

commit提交

Agent Skill

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

404

周安装

17

GitHub Stars

9

下载量

141
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:commit(提交)
来源仓库:https://github.com/kriscard/kriscard-claude-plugins
仓库路径:skills/commit
安装命令:
npx skills add https://github.com/kriscard/kriscard-claude-plugins --skill 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/kriscard/kriscard-claude-plugins --skill commit

简介

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 中围绕仓库状态、代码变更或协作事项进行整理。
  • 使用时需要结合来源仓库、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或文件读写。
  • 涉及命令执行时,应在提示词中明确确认步骤和失败处理方式。

SKILL.md

Git Commit Workflow

Direct execution of git commit workflow - no agent delegation, fast and simple.

Trigger Phrases

This skill activates on:

  • "commit changes", "commit this", "commit my work"
  • "save to git", "save my changes", "save this to git"
  • "push this", "push my changes", "push my work"
  • "create a commit", "make a commit"
  • "git commit", "check in my changes"

Proactive Triggering

After completing implementation work, proactively offer to commit:

  • "I've finished implementing the feature. Would you like me to commit these changes?"
  • Use AskUserQuestion with options: "Yes, commit now" / "No, I'll review first"

Workflow

Step 1: Assess Changes

git status
git diff

If no changes: Report "No changes to commit" and end.

Step 2: Stage Files

Stage logically related changes:

git add <files>

Skip sensitive files:.env, credentials, tokens - warn user if detected.

Step 3: Create Conventional Commit

Analyze changes and determine:

  • Type: feat, fix, docs, style, refactor, perf, test, chore
  • Scope: Component/area affected
  • Subject: Imperative, lowercase, max 50 chars, no period

Commit format:

<type>(<scope>): <subject>

[Narrative body explaining WHAT and WHY - 2-4 sentences, NO bullet points]

Good example:

feat(auth): add session timeout handling

Implements automatic session refresh when user activity is detected
within the timeout window. Sessions now persist across page reloads
using localStorage with encrypted tokens.

Bad example (avoid):

feat(auth): add features

- Added timeout
- Added refresh
- Added localStorage

Step 4: Execute Commit

git commit -m "$(cat <<'EOF'
<commit message here>
EOF
)"

Step 5: Handle Pre-commit Hooks

If hooks modify files:

  1. Stage the modified files
  2. Amend the commit: git commit --amend --no-edit

Step 6: Ask About Push

Use AskUserQuestion:

  • "Commit successful! Push to remote?"
  • Options: "Yes, push now" / "No, keep local"

Step 7: Push (if confirmed)

git push
# or if no upstream:
git push -u origin <branch>

Gotchas

  • Pre-commit hooks that modify files (formatters, linters) require re-staging — after hook failure, stage modified files and create a NEW commit (don't amend, as that modifies the previous commit)
  • git add. can accidentally stage unrelated files — prefer staging specific files by name
  • Claude tends to write commit bodies as bullet lists — use narrative prose (2-4 sentences) instead
  • When no upstream branch exists, git push fails silently — must use git push -u origin <branch>
  • Never amend commits already pushed to shared branches — this rewrites history others may have pulled

Safety Rules

  • NEVER commit.env files or credentials
  • NEVER force push without explicit user request
  • NEVER amend commits on shared branches without warning
  • Always show what will be committed before committing

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

36.2%
按下载量换算51

Claude

27.18%
按下载量换算38

Cursor

17.37%
按下载量换算24

Gemini CLI

9.62%
按下载量换算14

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

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