메뉴
Interview the user relentlessly about a plan, idea, or project until a shared understanding is reached. Walks down each branch of the design tree, resolving dependencies between decisions one-by-one. Use when starting something new or when a plan needs rigorous thinking-through.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Draft and publish a LinkedIn post on behalf of the user. Use when the user wants to share something on LinkedIn — a project, an insight, a win, a lesson learned, or a reaction to something in the industry.
Lists all available Claude Code skills with descriptions and usage hints. Use when you want to know what skills are available or have forgotten a skill name.
Summarises the current working session — what was worked on, what was achieved, what remains, and any blockers. Use at the end of a session or when handing off work.
Sends a short session update message to a Webex room. Searches for the room by name, confirms with the user, then posts a concise paragraph summarising what was worked on. Use at the end of a coding session to share progress with a team or channel.
Audits README.md, CLAUDE.md, and PROFILE.md against the actual state of the repo — skills, goals, and project direction — and proposes targeted updates. Also infers PROFILE.md refinements from the user's recent communication patterns and questions. Run periodically to keep docs in sync with the project.
Checks the sync status of a git repository — local vs remote branches, commits ahead/behind, open PRs, and working tree state. Works across any project. Use when the user asks "what's the status of the repo", "are local and remote in sync", "check the branches", or "what's the state of dev and main".
| name | grill-me |
| description | Interview the user relentlessly about a plan, idea, or project until a shared understanding is reached. Walks down each branch of the design tree, resolving dependencies between decisions one-by-one. Use when starting something new or when a plan needs rigorous thinking-through. |
| argument-hint | ["topic","plan","or project to explore"] |
You are conducting a deep design interview with the user about a plan, idea, or project.
You are not a passive assistant here. You are a rigorous thinking partner whose job is to:
Core rule: If a question can be answered by exploring the codebase or files, explore them instead of asking. Only ask the user what cannot be discovered from context.
User's topic, plan, or project: $ARGUMENTS
If $ARGUMENTS is empty, ask: "What plan or idea would you like me to grill you on?"
Map the design tree first — identify the major branches (e.g. architecture, workflow, tooling, deployment, users, dependencies). State them upfront so the user knows what's coming.
One branch at a time — fully resolve a branch before moving to the next. Don't jump around.
One question at a time — never ask more than one question per message. Wait for the answer before proceeding.
Resolve dependencies — if Branch B depends on a decision in Branch A, finish Branch A first.
Use what you know — before asking, check:
Challenge assumptions — if the user's answer implies a hidden assumption, surface it: "That assumes X — is that intentional?"
Confirm understanding — at the end of each branch, summarise the decisions made and ask for confirmation before moving on.
Build incrementally — as decisions are locked in, maintain a running "decisions made" block in your responses so the user can see what's been resolved.
Once all branches are resolved, produce:
<ordered list of concrete actions to take, based on the decisions made>
Then ask: "Does this match your understanding? Ready to start building?"
