agentleFS
Sign inStart free
AhmedMerzaAhmedMerza/agent-skills/skills/grill-me/SKILL.md

grill-me

Skill2 starsChanged 41 days agoMITView on GitHub

Interview the user relentlessly about a plan or design until shared understanding — working the design tree in rounds, asking every question whose prerequisites are already settled. Trigger ONLY on an explicit request to be grilled - /grill-me, "grill me", "poke holes in this", "stress-test this plan", "interrogate me about this". Also runs when /scout or /lead (world mode) reaches its roadmap gate with the user present - that hand-off is the request. Otherwise never infer it from a plan that merely looks underspecified, and never start it mid-task on your own judgment; it interrupts the work by design. Adapted from Matt Pocock's grill-me / grilling skills (github.com/mattpocock/skills).

We split this skill into the rules it gives an agent

3 rules. 0 are common, 2 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 178 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Ask the whole frontier, then wait for my answers before starting the next round.

    line 11
  2. Very specific:

    Ask each round through AskUserQuestion — one call, up to 4 questions, each with your recommended answer listed first and marked …

    line 18
  3. Distinctive:

    Finding facts is your job, never mine: When a frontier question needs a fact from the codebase or the environment, dispatch a subagent to …

    line 27

“Common” is measured: kept by at least 6 in 10 of the 178 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Interview me relentlessly until we reach a shared understanding. Map the plan 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, then wait for my answers before starting the next round.

A question whose answer depends on another question still open in this round belongs to a later round, not this one. That rule is what makes batching safe rather than merely faster: everything asked together is genuinely independent, so no answer in a round can invalidate another question inside it.

Ask each round through AskUserQuestion — one call, up to 4 questions, each with your recommended answer listed first and marked "(Recommended)". If the frontier is wider than 4, ask the 4 that unblock the most downstream decisions and carry the rest into the next round. Anything that doesn't fit that tool's shape — genuinely open-ended, needs a paragraph of context — ask in plain text alongside the call.

Each round's answers reshape the tree: settled decisions push the frontier outward and unblock questions that were waiting on them. Recompute the frontier and ask the next round.

Finding facts is your job, never mine. When a frontier question needs a fact from the codebase or the environment, dispatch a subagent to find it — never ask me something you could look up yourself. Don't block on it either: a running exploration is just an unsettled prerequisite, so only the questions downstream of it wait. Ask the rest of the frontier now.

The decisions, though, are mine. Put each one to me and wait.

The session is done when the frontier is empty — every branch of the tree visited, nothing left silently assumed. Don't act on the plan until I confirm we've reached shared understanding.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free