grill-me
Structured questioning to reach shared understanding before action. Walks the decision tree one branch at a time, checks memory for already-answered questions, saves every answer as a lesson. Use when user wants to think through a plan, align on approach, poke holes in a design, or figure out decisions before committing. Triggers include "before we start", "think this through", "what am I missing", "poke holes", "help me think/decide", "let's align", "interview me", "grill me", "challenge this", "what should I consider", "what would you ask".
- License 许可
- MIT
- 最近更新
- 2026-11-11
- Stars
- ★ 0
- Forks
- 0
01 前言 Front Matter
- name
- grill-me
- description
- Structured questioning to reach shared understanding before action. Walks the decision tree one branch at a time, checks memory for already-answered questions, saves every answer as a lesson. Use when user wants to think through a plan, align on approach, poke holes in a design, or figure out decisions before committing. Triggers include "before we start", "think this through", "what am I missing", "poke holes", "help me think/decide", "let's align", "interview me", "grill me", "challenge this", "what should I consider", "what would you ask".
02 资源文件 1
- 参考文档agents/openai.yaml122 B
03 SKILL.md 正文
Grill Me
HARD RULE: One Question Per Turn Your response MUST contain exactly ONE question. No exceptions. No "also" or "a few things." One question → wait → next question. Self-check: if you count more than one ? outside of quotes, delete the extras.
Activation
Explicit triggers ("grill me", "interview me", "challenge this"): Start immediately with the banner then ONE question:
🔥 Grill Mode — one concern at a time. Say "enough" when ready to move on.
Ambiguous triggers ("poke holes", "what am I missing", "think this through", "help me decide", etc.): Ask first:
One at a time, or full critique dump? [OPTIONS: Grill me one at a time | Just give me the full critique]
Turn Structure
Every grill turn = exactly this:
- Context (1-2 sentences) — why this matters
- Question — one decision only, ending with
? - My recommendation — what I'd pick and why (one line)
Rules
- Facts vs Decisions: Look up facts silently (code, config, memory). Only ask about decisions the user must make.
- Memory: Check lessons/memory first. If already decided, confirm: "Previously you decided X. Still holds?"
- Save answers:
learn_add(rule="<decision>", category="knowledge", scope="workspace")after each. - Document decisions: After 3+ decisions, offer once to capture them in a doc.
- Exit: On "enough"/"just do it" → summarize decisions, proceed to action. Don't implement mid-grill.
- Simple plans: If the plan is clear and simple, say so — don't manufacture questions.
04 同仓库更多 Skills
shell-script-expert
Use when the user asks "review this script", "write a bash script", "fix my shell script", "make this script safer", "why does this script fail", "convert this to a script", or working with .sh/.bash files, sh/bash syntax, shellcheck, cronjobs, and bash one-liners.
prose-synthesize
Turn unstructured notes into a structured, actionable plan. Use when given brain dumps, stream-of-consciousness, or scattered thoughts needing order.
prose-distill
Condense verbose text to its essence. Use when asked to condense, tighten, shorten, reduce verbosity, or omit needless words while preserving substance.
mindcraft
暂无简介
learn
Tutor for intellectual understanding: teach how or why something works rather than complete a task or provide an evaluative verdict. Trigger for explicit learning requests (teach, explain, ELI5, walkthroughs, quizzes, flashcards, definitions); terse concept prompts; confusion signals; prerequisite or sequencing questions; and conceptual questions about mechanisms, causes, or dynamics. Do not trigger for task completion (coding, writing, calculation, or translation), factual or news lookup, personal troubleshooting, resource recommendations, or opinion and interpretive verdicts.
deployment-readiness
Use to check deployment readiness of a host.