# Grill Me

> Use when the user asks to be grilled or wants one-question-at-a-time stress testing of a plan or design. Not for writing the final spec or implementing it; use scope or implement.

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

## Install

```sh
agentstack add skill-mblauberg-provenant-grill-me
```

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

## About

Interview the highest-leverage unresolved branch first. Resolve dependent
decisions one at a time; do not exhaustively interrogate low-value detail. For
each question, give the evidence already known, 2–3 concrete options when useful,
and a recommended answer with its trade-off.

Keep the explicit decision context current: intake/revision, constraints,
evidence, decided branches, parked owner calls and the next unresolved branch.
Do not start a parallel interview when revising the same request.

Ask one question at a time. Record decided, parked and still-open branches in
the enclosing `scope` artifact when write authority exists; otherwise keep a
compact chat register.

If a question can be answered from the workspace or current authoritative
sources, investigate it instead of spending user attention. Stop when the
shared decision/acceptance contract is clear, the user asks to stop, or the next
branch needs unavailable owner authority. Return the resolved register to
`scope` for a digest-bound handoff; the interview itself is not approval.

## Adapter-absent path

Console, Herdr and GitHub are optional. Continue from canonical project
artifacts and emit the skill-owned artifact kind in
[portable-workflow.v1.json](portable-workflow.v1.json). That filesystem
artifact records workflow evidence; it never approves the decision context.

## Source & license

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

- **Author:** [mblauberg](https://github.com/mblauberg)
- **Source:** [mblauberg/provenant](https://github.com/mblauberg/provenant)
- **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-mblauberg-provenant-grill-me
- Seller: https://agentstack.voostack.com/s/mblauberg
- 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%.