review-bench/asjas_course-platform/grill-me

Claude Code and agent skill by review-bench/asjas_course-platform

Trust tier 3
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

Interview the user relentlessly about a plan or design until reaching shared understanding. Use when: stress-test a plan, design review, grill me, poke holes, challenge assumptions, walk the decision tree, resolve design trade-offs.

Install

npx vskill@latest install review-bench/asjas_course-platform/grill-me

SKILL.md

Grill Me — Design Interview

Conduct a rigorous, structured interview about a plan or design. Walk down every branch of the decision tree, resolving dependencies between decisions one by one, until reaching shared understanding.

When to Use

  • The user has a plan, RFC, design doc, or architecture proposal.
  • The user wants assumptions challenged before implementation.
  • The user says "grill me", "poke holes", or "stress-test this".

Procedure

1. Understand the Plan
  • Read the plan the user provides (file, message, or selection).
  • If the plan references existing code, explore the codebase to build

context. Use subagents for large explorations.

  • Summarize your understanding back in 3-5 bullet points and ask

the user to confirm or correct.

2. Map the Decision Tree

Identify every decision point in the plan. Group them into branches:

  • Data model: schemas, relationships, migrations
  • API surface: endpoints, contracts, error handling
  • State & sync: client state, caching, real-time, offline
  • Security: auth, authz, input validation, secrets
  • UX: flows, edge cases, error states, accessibility
  • Operations: deployment, rollback, monitoring, migration path
  • Dependencies: libraries, services, infrastructure

List the branches explicitly so the user sees the full scope.

3. Walk Each Branch

For each branch, depth-first:

  1. Ask one focused question about the most critical or ambiguous

decision in that branch.

  1. Provide your recommended answer with brief rationale.
  2. If a question can be answered by exploring the codebase, explore

it yourself and present findings instead of asking.

  1. Wait for the user's response before moving on.
  2. If the answer reveals a new sub-decision, drill into it immediately

before continuing to the next branch.

  1. When a branch is fully resolved, summarize the decisions made and

move to the next branch.

4. Resolve Cross-Branch Dependencies

After walking all branches, identify conflicts or dependencies between decisions made in different branches. For example:

  • A data model choice that conflicts with the caching strategy.
  • An API design that doesn't support the offline-first requirement.
  • A security constraint that affects the UX flow.

Raise each conflict as a question with a recommended resolution.

5. Converge

When all branches and cross-dependencies are resolved:

  1. Present a decision summary — a compact list of every resolved

decision, grouped by branch.

  1. Highlight any open items that need further investigation or

are blocked on external factors.

  1. Ask the user to confirm the summary or reopen any decision.

Question Style

  • One question at a time. Never dump a list of 10 questions.
  • Be direct and specific: "How will you handle X when Y happens?"

not "Have you thought about error handling?"

  • Challenge weak answers: "That works for the happy path, but what

about [edge case]?"

  • Acknowledge strong answers briefly and move on.
  • If the user says "you decide" or defers, commit to your

recommendation and move forward.

Completion Criteria

The interview is complete when:

  • [ ] Every identified branch has been walked.
  • [ ] Cross-branch dependencies have been resolved.
  • [ ] The user has confirmed the decision summary.
  • [ ] Open items (if any) are explicitly listed.

Security

  • Security scanPassed, 95 of 100Pattern checks on the skill's files.
  • Model reviewPassed

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

Source

Popularity

7-day trend score
0

Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/review-bench/asjas_course-platform/grill-me/badge)](https://verified-skill.com/skills/review-bench/asjas_course-platform/grill-me)

Something wrong with this skill or its page? Tell us.