agentleFS
Sign inStart free
arndvsarndvs/ctrlshft/skills/grill-me/SKILL.md

grill-me

Skill0 starsChanged 6 months agoMITView on GitHub

Interview the user relentlessly about a plan or design until reaching shared understanding. Use when asked to 'grill me', 'interview me', 'ask me questions about this', or before writing a PRD to flesh out vague ideas.

We split this skill into the rules it gives an agent

8 rules. 0 are common, 6 are distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 178 edited versions we read keep these.

  • 6

    Distinctive

    This author's own choices. Worth a look.

  • 2

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    If running interactively (human present), output "Read Grill Me skill." to acknowledge.

    line 8
  2. Distinctive:

    Ask the questions one at a time, waiting for feedback before continuing.

    line 15
  3. Distinctive:

    If a question can be answered by exploring the codebase, explore the codebase instead.

    line 17
  4. Distinctive:

    When the user uses vague or overloaded terms, propose a precise canonical term.

    line 22
  5. Distinctive:

    When domain relationships or flows are being discussed, stress-test them with specific scenarios.

    line 25
  6. Distinctive:

    When the user states how something works, check whether the code agrees.

    line 28
  7. Distinctive:

    After reaching shared understanding, offer the user three paths:

    line 33
  8. Very specific:

    If context fills up during the interview, follow the standard handoff protocol (@~/dotfiles/instructions/handoff.instructions.md) — persist …

    line 41

“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

Grill Me

If running interactively (human present), output "Read Grill Me skill." to acknowledge. If running with --dangerously-skip-permissions (AFK/unattended), skip acknowledgement and proceed directly.

Pipeline position: /grill-me → /write-a-prd → /architect → /prd-to-issues → /do-work → shft

<what-to-do> Interview me relentlessly about every aspect of this plan until we reach a shared understanding. Walk down each branch of the design tree, resolving dependencies between decisions one-by-one. For each question, provide your recommended answer.

Ask the questions one at a time, waiting for feedback before continuing.

If a question can be answered by exploring the codebase, explore the codebase instead. </what-to-do>

<interviewing-discipline> ### Sharpen fuzzy language When the user uses vague or overloaded terms, propose a precise canonical term. "You're saying 'account' — do you mean the Customer or the User? Those are different things."

Discuss concrete scenarios

When domain relationships or flows are being discussed, stress-test them with specific scenarios. Invent edge cases that force the user to be precise about boundaries and behaviour.

Cross-reference with code

When the user states how something works, check whether the code agrees. If you find a contradiction, surface it: "Your code cancels entire Orders, but you just said partial cancellation is possible — which is right?" </interviewing-discipline>

Handoff

After reaching shared understanding, offer the user three paths:

  1. /write-a-prd — capture decisions as a formal PRD
  2. /prd-to-issues — break directly into GitHub issues
  3. /do-work — start implementing immediately

Let the user choose.

If context fills up during the interview, follow the standard handoff protocol (@~/dotfiles/instructions/handoff.instructions.md) — persist decisions made so far to working/ and provide the pickup command.

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.

Reviews can't be read right now.

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