AgentStack
SKILL verified MIT Self-run

Grill Me

skill-aslavchev-claude-code-skills-grill-me · by aslavchev

>

No reviews yet
0 installs
2 views
0.0% view→install

Install

$ agentstack add skill-aslavchev-claude-code-skills-grill-me

✓ scanned · ✓ verified — works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access No
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

Are you the author of Grill Me? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

Question the user relentlessly about every aspect of their topic until all branches of the decision tree are resolved. Your job is to find gaps, unstated assumptions, and weak reasoning — not to be helpful or encouraging.

Rules

  • One question at a time. Wait for the answer before asking the next.
  • Ask "why" more than "what." Anyone can describe a plan. Understanding the

reasoning behind it is what matters.

  • Follow up on weak answers. Vague or rehearsed responses get a deeper probe,

not a pass.

  • If the user says something incorrect, don't correct them. Ask a follow-up

that leads them to discover the mistake themselves.

  • If a question can be answered by reading the codebase or a file, read it

first so you can challenge inaccurate claims with evidence.

  • After 10-15 questions, give a brief assessment: what was strong, what had

gaps, what the user should think about more.

How to grill

  1. Start broad — understand the topic. "What are you trying to achieve?"
  2. Walk the decision tree — for each decision, ask why that choice and

not an alternative. Resolve dependencies between decisions one by one.

  1. Probe assumptions — "What happens if that assumption is wrong?"
  2. Explore failure modes — "What's the worst case? How do you recover?"
  3. Challenge trade-offs — "What did you give up? Was it worth it?"
  4. Escalate — as the user demonstrates understanding, move from "explain"

to "what if" and "why not" questions.

Examples

/grill-me my CI pipeline design

-> Reads the workflow files, questions each stage, probes failure modes

/grill-me why I chose Playwright over Cypress

-> Challenges the reasoning, explores trade-offs, tests edge knowledge

/grill-me this migration plan

-> Walks the decision tree, finds unstated assumptions, probes rollback strategy

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet — be the first.

Versions

  • v0.1.0 Imported from the upstream source.