Grilling — Agent Skill by Matt Pocock

From the author

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.

— the description by Matt Pocock

AutoGineer summary

Written by AutoGineer from the skill’s own files, not by the author

The grilling skill transforms the agent into a relentless interviewer designed to stress-test plans, ideas, or decisions. It operates by constructing a design tree where every decision branches into subsequent dependencies. The agent manages this process in rounds, identifying the current frontier of questions that can be answered immediately. It provides recommended answers for each question, phrased so that a 'yes' confirms the recommendation, and it proactively delegates fact-finding tasks to sub-agents to minimize user burden. The process continues iteratively until the entire tree is explored and a shared understanding is reached.

What it does

Maps user ideas or plans into a structured design tree of decisions.
Identifies the current frontier of questions that are ready to be answered.
Provides recommended answers for each question to streamline the decision-making process.
Uses sub-agents to fetch necessary facts from the environment instead of asking the user.
Iteratively updates the tree based on user input to uncover new dependencies.
Ensures all branches are explored before confirming a shared understanding.

When to use it

When you need to stress-test a complex plan or decision.
When you want to identify hidden assumptions in your reasoning.
When you need a structured, step-by-step breakdown of a project's requirements.
When you want to ensure no aspect of a decision has been overlooked.

Advantages

Forces rigorous examination of every decision point.
Reduces user effort by providing recommended answers and automating fact-finding.
Prevents oversight by mapping out all downstream dependencies.
Maintains a clear, logical progression through complex topics.

Usage examples

“Grill me on my plan to migrate our database to a new cloud provider.”: The agent begins by mapping out the migration plan, asking the first round of questions regarding current infrastructure and constraints, providing recommendations for each.
“I have an idea for a new marketing campaign, grill me on it.”: The agent initiates a structured interview, breaking the campaign down into target audience, budget, and channel decisions, asking questions round-by-round.

Install

Download the skill as a ZIP:
https://mp-staging.autogineer.com/agent-skills/mattpocock/grilling/download
Unzip it into the cross-client skills folder of your project:
```
mkdir -p .agents/skills && unzip grilling.zip -d .agents/skills
```

Documentation by Matt Pocock

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: 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.

Format a round like so:

❓ **Q1** - **<question title>**: <question body, might be multiple paragraphs, including multiple choices>

➡️ <your recommended answer>

---

❓ **Q2** - **<question title>**: <question body, might be multiple paragraphs, including multiple choices>

➡️ <your recommended answer>

Word each question so "yes" accepts your recommended answer.

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.), dispatch a sub-agent with a fresh context window to find it; 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; 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. Do not act on it until the user confirms you have reached a shared understanding.

Files

ATTRIBUTION.mdAdded by AutoGineer858 B
LICENSELicence file added by AutoGineer1.0 KiB
SKILL.md2.0 KiB
agents/openai.yaml113 B

LICENSE

MIT License

Copyright (c) 2026 Matt Pocock

Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.

Requirements and compatibility

Works with: Claude Code, Claude apps, OpenAI Codex, GitHub Copilot, Cursor, Gemini CLI, Kiro, OpenCode, Goose, Amp, Roo Code, JetBrains Junie, Factory Droid, Devin (ex-Windsurf), Mistral Vibe, OpenHands, Letta

“Works with” means the tool reads the open Agent Skills format; a skill written for one tool may use features others lack.

Security

No issues found by our static scan· No issues found

A static scan is not a guarantee. Skills are third-party content: review a skill before you let an agent use it.

Discernment Nudge
AnthropicOfficial
A framework for AI agents to append reflective follow-up questions to substantive, high-stakes responses.
Agent toolingClaude CodeClaude appsOpenAI Codex+14 more
Apache-2.0GitHub stars of the source repository188.6k
Wayfinder
Matt Pocock· Community
A structured approach to planning and executing complex, multi-session projects using a shared map of decision-based tickets on an issue tracker.
Agent toolingClaude CodeClaude appsOpenAI Codex+14 more
MITGitHub stars of the source repository317.5k
Brainstorming
Jesse Vincent (obra)· Community
A mandatory pre-work framework for clarifying user intent, exploring design approaches, and securing approval before implementation.
Agent toolingClaude CodeClaude appsOpenAI Codex+14 more
MITLow riskHas scriptsGitHub stars of the source repository314k
Interview Me
Addy Osmani· Community
A structured interview process to extract and verify a user's true intent before starting any work.
Agent toolingClaude CodeClaude appsOpenAI Codex+14 more
MITGitHub stars of the source repository113.9k
Idea Refine
Addy Osmani· Community
A structured ideation partner that refines raw concepts into actionable markdown one-pagers.
Agent toolingClaude CodeClaude appsOpenAI Codex+14 more
MITLow riskHas scriptsGitHub stars of the source repository113.9k
Doubt Driven Development
Addy Osmani· Community
A rigorous in-flight adversarial review process to stress-test non-trivial decisions before they are finalized.
Agent toolingClaude CodeClaude appsOpenAI Codex+14 more
MITGitHub stars of the source repository113.9k