grilling
Grill the user relentlessly about a plan, decision, or idea. Use when the user wants to stress-test their thinking, or uses any 'grill' trigger phrases.
Resources
1Install
npx skillscat add tt-a1i/matt-skills-with-to-goal/grilling Install via the SkillsCat registry.
Interview the user relentlessly until you reach a shared understanding. Map this as a design tree: every decision branches into the decisions that hang off it.
Work the tree in rounds. The frontier is every decision whose prerequisites are already settled โ the questions you can ask now without guessing at answers you haven't heard yet. Ask the whole frontier in one round: number each question and give your recommended answer. Then wait for the user's answers before the next round.
Question format
Lead each round with a fixed header, then emit every frontier question in one shape. Use emoji as structure anchors and question-type signals โ not decoration in every sentence.
Round header:
๐ฅ **Round N** ยท K questionsEach question (one type emoji before the number):
<type> **QN** - **<question title>**
<body โ prose and/or multiple choices>
๐ก <your recommended answer>Round footer (after the last question):
๐ Reply by number: `1) โฆ 2) โฆ 3) โฆ`Type emoji (pick one that fits the decision)
| Type | Emoji | Use when |
|---|---|---|
| Scope / in-or-out | ๐ฏ | Boundary, YAGNI, what ships now |
| Trade-off | โ๏ธ | A vs B, pick one path |
| Risk / failure | โ ๏ธ | What breaks, abuse cases, rollback |
| Naming / language | ๐ท๏ธ | Terms, glossary, how we say it |
| UX / UI | ๐ | What someone sees or clicks |
| Data / state | ๐งฉ | Model shape, transitions, schema |
| Cost / perf | โก | Latency, money, scale |
| Auth / trust | ๐ | Permissions, tenancy, secrets |
| Timing | โฑ๏ธ | When, sequencing, milestones |
| Challenge | ๐ช | "Do you actually need this?" |
| Dependency | ๐ | Blockers, prerequisites |
| Open / foggy | ๐ญ | Still ill-formed; need a direction |
Fixed skeleton (do not invent alternatives each round): ๐ฅ round ยท type emoji + QN ยท ๐ก recommendation ยท ๐ reply prompt.
Keep the body text clean โ at most one optional mood emoji in the round header line if it helps (e.g. a long session), never a spray of emoji inside the question body.
Round loop
Each round the user answers reshapes the tree โ settled decisions push the frontier outward and unblock questions that depended on them. Recompute the frontier and ask the next round. A question whose answer depends on another question still open in this round belongs to a later round, not this one.
Finding facts is your job, never the user's. When a frontier question needs a fact from the environment (filesystem, tools, etc.), dispatch a sub-agent to find it โ don't ask the user for anything you could look up yourself. Don't block on it: a running exploration is an unsettled prerequisite, so only the questions downstream of it wait for the sub-agent to report โ ask the rest of the frontier now. The decisions are the user's โ put each to them and wait.
The session is done when the frontier is empty: every branch of the design tree visited, nothing left silently assumed. Do not act on it until the user confirms you have reached a shared understanding.
Install
npx skillscat add tt-a1i/matt-skills-with-to-goal/grilling Install via the SkillsCat registry.