Skip to content
cccskills
FreePublished on GitHub

grilling

Stress-test a plan, decision, or idea as a dependency-aware design tree. Use for grill-me requests, assumption audits, and decision clarification before implementation.

The English summary is being prepared. Showing the original description.

mangowhoiscloud/geode
See how to install

Files included (1)

  • SKILL.md2.4 KB

SKILL.md (original)

Review the instructions the agent will receive before you install.

Grilling

Mode: decision clarification before action.

Treat $ARGUMENTS as the root decision. Build a compact design tree whose nodes are unresolved user decisions and whose edges are prerequisites. The current frontier is every unresolved node whose prerequisites are settled.

The runtime creates a typed grill_state before this prompt. Call update_grill(action="define") before presenting questions. The tree must be acyclic, contain 1-24 unique nodes, give every node 2-3 unique options, and name one recommended option with a reason. Prose is not state: only update_grill can settle a frontier answer or complete the interview. Use get_grill after a rejected update instead of guessing why the validator failed.

Loop

  1. Inspect available code, files, and sources for facts. Never ask the user for a fact that can be verified directly.
  2. Consider multiple plausible tree shapes internally. Keep the smallest tree that covers materially different outcomes; do not expose hidden reasoning.
  3. Ask every independent frontier question in one round. Defer a question when its answer depends on another unresolved node.
  4. For each question, state the decision, 2-3 mutually exclusive options, the consequence of each, and one recommended answer with a concrete reason.
  5. Wait for the user's answers. Update the tree, surface contradictions, and repeat until no unresolved frontier remains.
  6. Record answers only for nodes reported in the typed frontier. Call update_grill(action="complete") only after the validator reports no unresolved nodes, then summarize settled decisions, assumptions, rejected branches, and open risks. Do not implement until the user confirms the shared understanding.

Use this shape for each frontier item:

Q1 — <decision>
- A: <consequence>
- B: <consequence>
Recommendation: <choice and reason>

Do not simulate MCTS or LATS. Shared files, shells, and external systems are not cloneable branch state. This is a decision-tree interview, not speculative parallel execution.

Reviews

No reviews yet. Tell others what you thought.

Comments

No login needed. Share how it worked for you, ask a question, or note where it did not run.

No comments yet. Be the first to write one.

Comments are published right away. Do not include personal information such as email addresses or phone numbers. Up to 2 URLs.

0 / 1,000 characters

This form is protected by reCAPTCHA, and the Google Privacy Policy and Terms of Service apply.

More skills from this repository

agent-anti-pattern

Free

Audit agent runtimes and their tests for lifecycle inference, capability overload, context pollution, correlated verification, prompt-only control, false success, dead compatibility paths, test-only seams, and complexity without contribution. Use for whole-repository agent audits, slop or dead-code triage, test contribution reviews, large refactors, tool/context simplification, and cleanup requests that need proof before deleting code or tests.

The English summary is being prepared. Showing the original description.

mangowhoiscloud/geode15Updated Oct 22, 2026

agent-ops-debugging

Free

Diagnose agent operational failures through default-policy checks, multi-gap analysis, explicit service ownership, request-context lifetime, and degradation-versus-correctness checks.

The English summary is being prepared. Showing the original description.

mangowhoiscloud/geode15Updated Oct 22, 2026

agent-world-benchmark

Free

Align, run, audit, and publish GEODE benchmark comparisons against Agent-World arXiv 2604.18292v1 across MCP-Mark, BFCL V4, and tau2-bench. Use for Agent-World or AgentWorld comparison requests, cross-suite agent benchmark design, Tau2 comparator/profile diagnosis, paired thin-runtime-versus-GEODE experiments, mean-accuracy@8 replication, or benchmark comparability and artifact reviews.

The English summary is being prepared. Showing the original description.

mangowhoiscloud/geode15Updated Oct 22, 2026

anti-deception-checklist

Free

Verification checklist to prevent fake success. Detects test deletion/disabling, coverage regression, lint bypass, secret exposure. Triggered by "deception", "fake" ("가짜"), "fake success", "verification" ("검증"), "checklist" ("체크리스트"), "deletion detection" ("삭제 탐지"), "regression" keywords.

The English summary is being prepared. Showing the original description.

mangowhoiscloud/geode15Updated Oct 22, 2026

architecture-patterns

Free

Implement proven backend architecture patterns including Clean Architecture, Hexagonal Architecture, and Domain-Driven Design. Use when architecting complex backend systems or refactoring existing applications for better maintainability.

The English summary is being prepared. Showing the original description.

mangowhoiscloud/geode15Updated Oct 22, 2026

arxiv-digest

Free

Auto-search and summarize latest AI/agent papers.

The English summary is being prepared. Showing the original description.

mangowhoiscloud/geode15Updated Oct 22, 2026

See all skills by mangowhoiscloud

Report a problem with this skill