alexion/dev
1
0 Fork 0
Code Issues Pull Requests 1 Actions Packages Projects Releases Wiki Activity

fix(grill): restore upstream interview flow #216

Merged
alexion merged 1 commits from fix/reimport-grill-skill into main 2026-10-30 10:35:38 -04:00
Conversation 0 Commits 1 Files Changed 1
+10 -28
1 changed files with 10 additions and 28 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/skills/grill/SKILL.md
+10 -28
View File
@@ -3,46 +3,28 @@ name: grill
description: 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.
---
Interview the user relentlessly until you reach a shared understanding. Map this as a **design tree**: each decision branches into the decisions it unblocks.
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.
Maintain the design tree in a new Markdown file. Use the file as the source of truth.
Work the tree in **rounds**. The **frontier** is every decision whose prerequisites are already settled: 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.
Build the complete **frontier** before the first question.
After each answer, recompute and display every frontier decision.
Ask one frontier decision per turn.
A clarification leaves that decision open.
Format a question like so:
Format a round like so:
```
❓ **D1 — <decision title>**: <question body, might be multiple paragraphs, including multiple choices>
❓ **Q1** - **<question title>**: <question body, might be multiple paragraphs, including multiple choices>
**Options**
1. <option>
...
<n>. **<recommended option>** — recommended
➡️ <your recommended answer>
---
❓ **D2 — <decision title>**: <question body, might be multiple paragraphs, including multiple choices>
❓ **Q2** - **<question title>**: <question body, might be multiple paragraphs, including multiple choices>
**Options**
1. <option>
...
<n>. **<recommended option>** — recommended
➡️ <your recommended answer>
```
List every viable answer in an ordered list and mark your recommended answer.
Word each question so "yes" accepts your recommended answer.
Before asking the next question, account for every decision that the response settles.
Record preferences as constraints.
Settle an implementation decision only after facts establish that it is viable.
A decision with unresolved prerequisites belongs outside the frontier.
Each round the user answers reshapes the tree: 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.), find it with available tools; don't ask the user for anything you could look up yourself. The _decisions_ are the user's: put each to them and wait.
Finding _facts_ is your job, never the user's. When a frontier question needs a fact from the environment (filesystem, tools, etc.), find it with available tools; 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 result; ask the rest of the frontier now. The _decisions_ are the user's: 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. Present every settled decision from the design tree. Do not act on it until the user confirms you have reached a shared understanding.
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.