菜单
Use when creating or improving agent skills - ensures efficient context usage, clear activation conditions, and proper structure for different skill types
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Use when implementing any task end-to-end without user intervention - ensures continuous forward progress by always having a next tool call until the task is fully complete
Use when a task should be parallelized across multiple agents — decomposes work into Linear issues, then drains the queue by spawning worktree agents for unblocked tickets using drive-to-completion
Use when creating git commits - ensures messages follow Conventional Commits, explain motivation over implementation, and help future maintainers
Use when creating a Linear issue from a feature, bug, or task description — gathers required fields then calls the Linear MCP tool directly
Use to critically review a plan before approval - invokes a senior engineering review for readability, maintainability, operability, and testability
Use when implementing any feature or bugfix, before writing implementation code - write the test first, watch it fail, write minimal code to pass; ensures tests actually verify behavior by requiring failure first
| name | Write Skills |
| description | Use when creating or improving agent skills - ensures efficient context usage, clear activation conditions, and proper structure for different skill types |
Context window is shared with system prompts, conversation history, and other skills. Keep skills under 150 lines, ideally under 100. Challenge each piece of information: do I truly need this explanation?
The description field clarifies activation conditions, not summaries.
description: This skill helps write git commit messages following best practices description: Use when creating git commits - ensures messages follow Conventional Commits, explain motivation over implementation, and help future maintainersMulti-step workflows need structure:
<required> blocks with explicit step-by-step instructions<system-reminder> tags for critical decision points</good-example>
## Integration Skills
CLI tools and scripts should do the work:
- Keep skill thin - delegate to external tools
- Rely on tool error messages to guide behavior
- Don't over-explain what the tool already handles
<bad-example>
Lengthy explanations of all CLI flags, error cases, and edge conditions
</bad-example>
<good-example>
Brief instruction to run CLI tool, let its help/errors guide the agent
</good-example>
# Formatting Conventions
- Use fake XML tags: `<good-example>`, `<bad-example>`, `<required>`, `<system-reminder>`
- Write first person: "Ask me to do ABC" (not "Ask the user")