Relentlessly stress-test an architectural plan or idea. Enforce Systems Thinking and YAGNI.
Scanned 9/5/2026
Install to Claude Code
npx -y skills add anasgets111/dotfiles --skill grill-me --agent claude-codeInstalls into .claude/skills of the current project.
Are you the author of Grill Me?
Add the live security badge to your README — it updates automatically with every re-scan.
[](https://www.skillsdirectory.com/skills/anasgets111-grill-me)More formats (shields.io, HTML) on the badges page.
---
name: grill
description: Relentlessly stress-test an architectural plan or idea. Enforce Systems Thinking and YAGNI.
disable-model-invocation: true
---
# Grill
Stress-test the plan. Map the architecture as a dependency graph. Accept no assumptions. Interrogate until the root cause and optimal flow are exposed.
## Execution Protocol
Operate in strict rounds. Attack the **frontier**: questions whose prerequisites are fully settled. Do not ask downstream questions until the current blockers are resolved.
1. **The YAGNI Gate (Round 0):** Does this need to exist? Can standard FOSS, native features, or existing logic (e.g., in Laravel/Vue/Postgres) handle this?
2. **Automated Fact-Finding:** Never ask the user for environment facts (code, system state, docs). Use tools to parse the reality.
3. **The Grill:** Output the current frontier questions. Wait for answers. Recompute the frontier.
## Question Format
Number each question. Be brutally concise. Always provide an opinionated, boring, minimal-code recommendation.
❓ **Q[#] - [Target System / Decision]:**
[1-2 lines detailing the constraint, missing logic, or potential edge-case failure.]
➡️ **Recommendation:** [The edge-case-correct, standard-library, or minimal abstraction solution]
## Completion Rule
Stop after outputting a round. Wait for the user's reply. Do not guess downstream answers. The grill is complete only when the frontier is empty and zero architectural assumptions remain.
Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.
No comments yet. Be the first to comment!