devton/agentic-workflow-blueprint/grilling
Claude Code and agent skill by devton/agentic-workflow-blueprint
Trust tier 3
- community
Source check pending
Checking that the SKILL.md file is still in the repository.
Deep Socratic grilling and interrogation protocol to stress-test assumptions, clarify non-functional requirements, and resolve edge cases before design.
Install
npx vskill@latest install devton/agentic-workflow-blueprint/grillingSKILL.md
agentic-workflows-blueprint.workflow.mattpocock.grilling
Goal
Systematically interrogate assumptions, edge cases, trade-offs, and operational boundaries through intensive Socratic questioning before domain modeling or code implementation.
Scope
- Applies to: Feature requirement clarification, ambiguous requests, architectural trade-offs, and high-impact changes.
- Does not cover: Direct code implementation or initial codebase mapping (use
/wayfinder).
Triggers
- "/grilling"
- "Grill user on feature details"
- "Challenge feature assumptions"
- "Socratic grilling session"
Inputs
wayfinderBrieforuserPrompt: Orientation brief or initial feature request.existingRootDoc: Main project instruction document (AGENTS.md,CLAUDE.md).
Invariants (Guardrails)
- Mandatory Socratic Interrogation Gate: Present 3–7 strategic questions covering Scope, Edge Cases, Data Integrity, Security, and Operational Constraints. Do NOT proceed without explicit user responses.
- Option-Driven Questioning: Present questions with clear trade-offs (Pros, Cons, Default Option) so choices are actionable.
- Strict UUIDv7 Ordering Invariant: Enforce time-sortable UUIDv7 for resource identifiers in data model decisions.
- Binding Decision Log: Record all user answers and trade-off resolutions into a binding decision log to govern domain modeling and spec creation.
Procedure
1) Analyze Wayfinder & Request Context
- Review the orientation brief (
wayfinder.md) and original request to extract ambiguous requirements, boundary decisions, and risk areas. - Identify implicit assumptions that must be explicitly confirmed.
2) Formulate Categorized Grilling Questions
Formulate 3 to 7 targeted questions across key dimensions:
- Category A: Scope & Boundary Limits — What is strictly in vs. out of scope for this iteration?
- Category B: Edge Cases & Failure Modes — How to handle network timeouts, concurrency conflicts, invalid payloads, or missing data?
- Category C: Data Contracts & Invariants — Permitted mutations, state transition guards, UUIDv7 identity rules, and audit logging.
- Category D: Performance & Security Constraints — Rate limiting, RBAC permissions, zero-downtime deployment requirements, and migration strategy.
Format each question with:
- Context & Problem Statement
- Option 1 (Pros / Cons)
- Option 2 (Pros / Cons)
- Recommended Default Option
3) Present Grilling Interview
- Present the formatted questions clearly to the user.
- Wait for explicit user answers. Do NOT proceed to domain modeling or spec writing until grilling responses are recorded.
4) Synthesize Binding Decision Log
- Summarize user choices and confirmed trade-offs into a binding Decision Log.
- Store decision log in conversation context and append to
walkthrough.md.
Outputs
- Binding
decision-log.mdstored in conversation context andwalkthrough.md.
Review gate
- [ ] At least 3 to 7 targeted grilling questions presented with trade-offs.
- [ ] Explicit user responses received and recorded.
- [ ] UUIDv7 identity standard confirmed for resource data models.
- [ ] Binding decision log generated to govern domain modeling and specification.
References
../../../SKILL.md../wayfinder/SKILL.md../domain-modeling/SKILL.md../../embed-aihero-radioactive/SKILL.md
Security
- Security scanPassed, 90 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingworkflows/mattpocock/grilling/SKILL.md
Popularity
- Repository stars
- 72
- Repository forks
- 4
- 7-day trend score
- 0
Stars and forks are for the whole GitHub repository, not this skill alone. 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
[](https://verified-skill.com/skills/devton/agentic-workflow-blueprint/grilling)