domains / software-engineering / plan-grilling-with-docs / grill-with-docs

grill-with-docs

Skill card

combined

The first copy is the canonical, minimal instruction: call the Skill tool twice for 'grilling' and 'domain-modeling'. The second copy is a catalog wrapper that adds operational constraints (inspect existing ADR/glossary conventions, record only confirmed answers, mark open proposals as Proposed, preserve conflicting alternatives, review the diff) and a conditional guard for domain-modeling. These are material steps the first copy lacks, so crowning the first would hide a real alternative procedure.

Repositories

RepositoryPath
mattpocock/skillsskills/engineering/grill-with-docs/SKILL.md
sickn33/agentic-awesome-skillsplugins/agentic-awesome-skills-claude/skills/grill-with-docs/SKILL.md

Other copies

RoleRepositoryPath
peermattpocock/skillsskills/engineering/grill-with-docs/SKILL.md
peersickn33/agentic-awesome-skillsplugins/agentic-awesome-skills-claude/skills/grill-with-docs/SKILL.md
adaptersickn33/agentic-awesome-skillsplugins/agentic-awesome-skills/skills/grill-with-docs/SKILL.md
adaptersickn33/agentic-awesome-skillsskills/grill-with-docs/SKILL.md

Combined skill

---
name: grill-with-docs
description: A relentless interview to sharpen a plan or design, which also creates docs (ADR's and glossary) as we go.
disable-model-invocation: true
---

## Steps

1. Inspect existing ADR and glossary conventions in the project before starting.
2. Use the installed `grilling` procedure to ask one meaningful question at a time.
3. Record only confirmed answers in authorized project docs. Mark open proposals as Proposed. Preserve conflicting alternatives.
4. Use the `domain-modeling` procedure only if it is installed and relevant to the current plan or design. Do not invent a slash-command dependency.
5. Review the resulting diff of the generated or updated docs before treating them as final.

## Limitations

- Requires the upstream tool, account, API key, or local setup when the workflow names one.
- Does not authorize destructive, production, paid, or external-message actions without explicit user approval.
- Validate generated artifacts or recommendations against the user's real sources before treating them as final.