# Grilling

> Stress-test a plan, decision, or idea through evidence and consequential tradeoffs. Use when the user asks to grill or challenge their thinking.

- **Type:** Skill
- **Install:** `agentstack add skill-waffleflopper-ai-tools-grilling`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [waffleflopper](https://agentstack.voostack.com/s/waffleflopper)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [waffleflopper](https://github.com/waffleflopper)
- **Source:** https://github.com/waffleflopper/ai-tools/tree/main/skills/grilling

## Install

```sh
agentstack add skill-waffleflopper-ai-tools-grilling
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

Investigate the relevant implementation, constraints, and recorded decisions before
asking questions. Resolve discoverable facts yourself. Reuse settled decisions;
reopen one only when new evidence creates a material conflict, explaining why.

Ask about unresolved choices with materially different product outcomes, risks,
or tradeoffs. Give a recommendation, its reason, and a concrete example. Challenge
weak assumptions candidly, including the user's preferred answer when warranted.

Choose routine engineering defaults and behavior already implied by accepted
requirements. Summarize those defaults together so the user can correct them;
do not turn correctness, consistency, or every internal design choice into an
approval question. For example, a derived forecast normally updates when its
source date is corrected; choosing which source governs that forecast may need
an owner decision.

Work through dependent or consequential questions one at a time. Batch independent,
straightforward questions when useful or requested. Follow the user's preferred
pace. Continue independent investigation and authorized preparation while an
answer is pending; do not implement behavior that depends on an unresolved choice.

Record the agreed outcome, decisions, routine defaults, and remaining blockers in
the existing work record when one is used. Finish planning with the next concrete
action and a concise continuation prompt when handing off; reference the record
instead of repeating instructions. Planning alone does not authorize implementation
or publication, but do not ask again for authorization already granted in the task.

## Source & license

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

- **Author:** [waffleflopper](https://github.com/waffleflopper)
- **Source:** [waffleflopper/ai-tools](https://github.com/waffleflopper/ai-tools)
- **License:** MIT

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

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-waffleflopper-ai-tools-grilling
- Seller: https://agentstack.voostack.com/s/waffleflopper
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.