菜单
Handles user confusion by explaining actual behavior and determining whether there is a real issue to address. Primary goal is clarification, not a fix. Use when the user says "I don't understand", "this doesn't make sense", "why does this do X", or shows contradictory expectations about existing code.
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
Prepares a git commit for react-gts-app by analyzing staged changes and detecting commit style from the repo's signals — user history, recent repo history, commit templates, and any commit-msg hook. Confirms with the user only when signals conflict. Use when the user wants to commit changes or asks to prepare a commit message.
Studies existing code patterns, architecture, and conventions in react-gts-app. Reads reference components and tests to extract naming, import order, RTL query priority, and GTS-enforced conventions before writing new code. Use when writing new code or entering an unfamiliar area.
Runs the appropriate linter and formatter for specified files in react-gts-app. Maps file extensions to GTS-backed tools (eslint, prettier) and reports remaining errors. Use when linting a specific file or directory, or after editing code before moving on.
Generates a production-ready AGENTS.md and complete .claude/ setup through guided codebase discovery and targeted interview. Detects the user's global setup, the project's filesystem, and installed tooling — then asks the user to confirm rather than assuming. Creates skills, hooks, settings, and symlinks tailored to the project. Use when setting up a new project for AI-assisted development or updating an existing setup.
Implements a new feature in react-gts-app using a dialogue-first workflow: researches the codebase, asks one clarifying question at a time, explores 2–3 approaches with trade-offs, validates the chosen design section-by-section, then implements after approval. Use when implementing new functionality or a significant change.
Guides collaborative dialogue to turn ideas into designs through one-at-a-time questions, approach exploration with trade-offs, and incremental validation. Does not implement code. Use before creative work or significant changes — when the user asks to "brainstorm", "think through", "explore options for", or similar.
| name | clarify |
| description | Handles user confusion by explaining actual behavior and determining whether there is a real issue to address. Primary goal is clarification, not a fix. Use when the user says "I don't understand", "this doesn't make sense", "why does this do X", or shows contradictory expectations about existing code. |
| user-invocable | true |
| disable-model-invocation | false |
Explain first. Do not jump to a fix.
Before reading anything, repeat back what you think the user is confused about in one sentence. If you're guessing, ask a clarifying question instead.
Open the exact files the user is looking at. Include tests (*.test.tsx) — they often document intent. Do not read more broadly than needed.
Describe the real behavior, not what the code "should" do. Reference specific lines with file:line. If the behavior differs from what the user expected, say why:
Use AskUserQuestion:
{
"questions": [{
"question": "Explanation clear. What's next?",
"header": "Next step",
"options": [
{"label": "Accept as-is", "description": "The current behavior is correct. Move on."},
{"label": "File as change request", "description": "The behavior is wrong or unclear. Open /add-feature to design a change."},
{"label": "Investigate further", "description": "Dig deeper into related code, tests, or history before deciding."}
],
"multiSelect": false
}]
}
file:line whenever you reference behavior./add-feature or a direct fix. Don't patch in place inside a clarify session.