grill-spec
Grill Spec
Before presenting a spec or design document to the user for review, offer to stress-test it with grill-me first. Grilling catches design blind spots — untested assumptions, missing edge cases, weak trade-off reasoning. Without this step, the user ends up doing the stress-testing themselves.
Steps
-
Check grill-me dependency — if grill-me is not installed, ask user if they want to install it (see references/install-grill-me.md for installation paths). If user declines, skip grilling entirely.
-
Ask and invoke — ask the user if they'd like to stress-test the spec with grill-me before their review. If user agrees, invoke grill-me with the spec file path. If user declines, proceed directly to user review.
-
Offer spec updates — when grilling completes, if the session produced new insights or decision changes, ask: "Want me to update the spec with any of the insights from this session?" If nothing new came up, skip this and proceed to user review.
Example
[Spec is ready — about to hand to user for review]
Agent: Before your review, want me to stress-test the spec with grill-me?
It catches untested assumptions and edge cases.
User: Sure, go ahead.
[Invokes grill-me with spec path: docs/design/auth-redesign.md]
[grill-me asks several questions, user answers each]
Agent: Grilling done. We uncovered a gap in the token rotation policy.
Want me to update the spec with that?
User: Yes, add it.
Agent: Done. Ready for your full review.
Scope
Applies to: standalone spec or design documents — e.g. files produced by writing-plans or brainstorming skills, docs/design/*.md, architecture decision records.
Does NOT apply to: code review, PR review, PR descriptions, commit messages, or design discussions embedded in conversation.
Skip Policy
The user can always skip grilling. If they seem pressed for time, briefly explain the value ("stress-tests design assumptions before your review"), then ask if they'd like to skip. Respect their answer — the user owns the process, not the skill.
More from shihyuho/skills
lessons-learned
Use when starting, executing, or finishing non-trivial implementation tasks where reusable constraints may matter. Recall relevant lessons before work, capture reusable corrections or discoveries during and after work, and keep project memory in `docs/lessons/`.
101fanfuaji
Use when user requests Chinese terminology conversion, checking, or ensuring terminology - "使用繁體中文", "使用台灣用語", "轉換成台灣用語", "確保都是台灣用語", "統一台灣用語", "改成台灣用語", "用台灣的說法", "簡體轉繁體", "繁體轉簡體", "全部改成繁體", "轉成台灣繁體", check/ensure Taiwan/Hong Kong/China terminology, simplified/traditional conversion, or phonetic transcription (Pinyin/Bopomofo)
60executing-plans-preflight
Use before any implementation start — auto-detects and fixes git state issues (branch, dirty files, remote sync) with one confirmation per fix. Trigger on "start implementation", "implement this plan", "start coding", "execute plan", "開始實作", "執行計劃", or any signal that coding is about to begin.
42agent-install-guide
Use when creating INSTALL.md, setup guides, or configuration instructions intended to be executed by AI agents.
35harvest
Capture project memory from planning-with-files source-of-truth into Obsidian-compatible second-brain notes in docs/notes. Use for milestone summaries, decision capture, and timeline snapshots. Trigger on: harvest, /harvest, harvest this, save this to second brain, save what we just did, document this work, capture this knowledge.
29skill-design
Design and refactor Agent Skills with concise, high-signal instructions and explicit trigger metadata. Use when creating a new skill, revising SKILL.md/README.md structure, or improving skill discoverability and portability.
27