AGENT SKILL

grill-with-docs

Use when the user asks to be grilled AND wants the design written down as you go — a relentless interview that sharpens terminology and records the domain model (glossary + ADRs) inline.

License 许可
MIT
所属仓库
jratlee/FDI
最近更新
2026-10-19
Stars
★ 0
Forks
0

01 前言 Front Matter

name
grill-with-docs
author
Matt Pocock (mattpocock/skills, MIT); adapted for Hermes
license
MIT
version
1.0.0
metadata
{ "hermes": { "tags": [ "design-review", "questioning", "domain-modeling", "glossary", "adr", "planning" ], "related_skills": [ "productivity/grilling", "engineering/domain-modeling", "productivity/grill-me" ] } }
description
Use when the user asks to be grilled AND wants the design written down as you go — a relentless interview that sharpens terminology and records the domain model (glossary + ADRs) inline.

02 资源文件 2

03 SKILL.md 正文

Grill With Docs

Adapted from mattpocock/skills -> skills/engineering/grill-with-docs (MIT, upstream body is a single line: Call the Skill tool twice, for "grilling" and "domain-modeling"). Hermes has no Skill tool, so the pointer is rewritten below. Behaviour is unchanged.

When to Use

  • The user says "grill me with docs", "grill me on this codebase", "stress-test this and write down the decisions", or asks for an interview where the output should be design documentation.
  • Use plain productivity/grill-me when they only want the interrogation and no docs.

Do this now

Load BOTH skills and work them together -- do not paraphrase from memory.

# Hermes
skill_view(name="productivity/grilling")
skill_view(name="engineering/domain-modeling")

# any other agent (Replit / Claude Code / Codex): read the files directly, wherever
# the skills root puts them (category folders vary by install)
<skills-root>/**/grilling/SKILL.md
<skills-root>/**/domain-modeling/SKILL.md

grilling supplies the interview discipline (rounds, the frontier, a recommended answer per question). domain-modeling supplies the writing discipline (challenge terms, sharpen fuzzy language, test scenarios, update the glossary and ADRs the moment a decision crystallises).

How the two combine

  • Each round is still a grilling round: numbered questions, one recommended answer each, wait for the answers before the next round.
  • While asking, apply domain-modeling inline: when an answer resolves a term, write it to GLOSSARY.md immediately; when a decision passes all three ADR tests (hard to reverse, surprising without context, a real trade-off), offer the ADR and write it then.
  • Before asking a question, check the code and the existing GLOSSARY.md / docs/adr/ first. A contradiction between what the user says and what the code does is a question, not a footnote.
  • Docs are the side effect of settled decisions, never a way to pre-empt them: never write a term or an ADR the user has not agreed to.

Hermes notes

  • Use clarify for pickable rounds (up to 5 per call); drop to numbered prose when a question needs a long answer.
  • Fact-finding is yours: read the repo, run commands, or delegate_task for wide searches. Never ask the user for something you could look up. Don't block the whole round on a running search -- only the questions downstream of it wait.
  • In a repo, the docs live at the repo root (GLOSSARY.md, docs/adr/, or GLOSSARY-MAP.md + per-context glossaries). Outside a repo (Obsidian, business planning), put the same two files beside the work you are planning and say where you wrote them.
  • Report at the end: which terms were added, which ADRs were written, and which questions are still open.

04 同仓库更多 Skills

jratlee
jratlee/FDI

vercel-react-native-skills

React Native and Expo best practices for building performant mobile apps. Use when building React Native components, optimizing list performance, implementing animations, or working with native modules. Triggers on tasks involving React Native, Expo, mobile performance, or native platform APIs.

★ 0
jratlee
jratlee/FDI

vercel-react-best-practices

React and Next.js performance optimization guidelines from Vercel Engineering. This skill should be used when writing, reviewing, or refactoring React/Next.js code to ensure optimal performance patterns. Triggers on tasks involving React components, Next.js pages, data fetching, bundle optimization, or performance improvements.

★ 0
jratlee
jratlee/FDI

vercel-composition-patterns

React composition patterns that scale. Use when refactoring components with boolean prop proliferation, building flexible component libraries, or designing reusable APIs. Triggers on tasks involving compound components, render props, context providers, or component architecture. Includes React 19 API changes.

★ 0
jratlee
jratlee/FDI

supabase-postgres-best-practices

Postgres performance optimization and best practices from Supabase. Use this skill when writing, reviewing, or optimizing Postgres queries, schema designs, or database configurations.

★ 0
jratlee
jratlee/FDI

skill-creator

Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.

★ 0
jratlee
jratlee/FDI

grilling

Use when the user wants a plan stress-tested.

★ 0