generating-smart-commits
关于
This skill automatically generates conventional commit messages by analyzing staged Git changes. It determines commit types, identifies breaking changes, and formats messages to standard specifications. Use it when creating commits via commands like `/gc` or after staging changes with `git add`.
快速安装
Claude Code
推荐/plugin add https://github.com/jeremylongshore/claude-code-plugins-plusgit clone https://github.com/jeremylongshore/claude-code-plugins-plus.git ~/.claude/skills/generating-smart-commits在 Claude Code 中复制并粘贴此命令以安装该技能
技能文档
Overview
This skill empowers Claude to create well-formatted, informative commit messages automatically. By analyzing staged changes, it generates messages that adhere to conventional commit standards, saving developers time and ensuring consistency.
How It Works
- Analyzing Staged Changes: The skill examines the changes currently staged in the Git repository.
- Generating Commit Message: Based on the analysis, it constructs a conventional commit message, including type, scope, and description.
- Presenting for Confirmation: The generated message is displayed to the user for review and approval.
When to Use This Skill
This skill activates when you need to:
- Create a commit message from staged changes.
- Generate a conventional commit message.
- Use the
/commit-smartor/gccommand. - Automate the commit message writing process.
Examples
Example 1: Adding a New Feature
User request: "Generate a commit message for adding user authentication"
The skill will:
- Analyze the staged changes related to user authentication.
- Generate a commit message like:
feat(auth): Implement user authentication module. - Present the message to the user for confirmation.
Example 2: Fixing a Bug
User request: "/gc fix for login issue"
The skill will:
- Analyze the staged changes related to the login issue.
- Generate a commit message like:
fix(login): Resolve issue with incorrect password validation. - Present the message to the user for confirmation.
Best Practices
- Stage Related Changes: Ensure that only related changes are staged before generating the commit message.
- Review Carefully: Always review the generated commit message before committing to ensure accuracy and clarity.
- Provide Context: If necessary, provide additional context in the request to guide the AI analysis (e.g.,
/gc - emphasize that this fixes a security vulnerability).
Integration
This skill integrates directly with the Git repository through Claude Code. It complements other Git-related skills by providing a streamlined way to create informative and standardized commit messages.
GitHub 仓库
相关推荐技能
content-collections
元Content Collections 是一个 TypeScript 优先的构建工具,可将本地 Markdown/MDX 文件转换为类型安全的数据集合。它专为构建博客、文档站和内容密集型 Vite+React 应用而设计,提供基于 Zod 的自动模式验证。该工具涵盖从 Vite 插件配置、MDX 编译到生产环境部署的完整工作流。
sglang
元SGLang是一个专为LLM设计的高性能推理框架,特别适用于需要结构化输出的场景。它通过RadixAttention前缀缓存技术,在处理JSON、正则表达式、工具调用等具有重复前缀的复杂工作流时,能实现极速生成。如果你正在构建智能体或多轮对话系统,并追求远超vLLM的推理性能,SGLang是理想选择。
evaluating-llms-harness
测试该Skill通过60+个学术基准测试(如MMLU、GSM8K等)评估大语言模型质量,适用于模型对比、学术研究及训练进度追踪。它支持HuggingFace、vLLM和API接口,被EleutherAI等行业领先机构广泛采用。开发者可通过简单命令行快速对模型进行多任务批量评估。
langchain
元LangChain是一个用于构建LLM应用程序的框架,支持智能体、链和RAG应用开发。它提供多模型提供商支持、500+工具集成、记忆管理和向量检索等核心功能。开发者可用它快速构建聊天机器人、问答系统和自主代理,适用于从原型验证到生产部署的全流程。
