Skip to content
SkillModel-led18 tokens in use

Grill With Docs

A user-invoked skill that runs the grilling interview and the domain-modeling discipline together, leaving a glossary and ADRs behind.

328,754 stars on the repository MIT Licensemattpocock/skills · skills/engineering/grill-with-docs
claude plugin install mattpocock-skills@claude-plugins-official
View source

Install command as written in the repository's README

Skill label

measured from its files

What installing Grill With Docs costs

31tokens idle
18tokens in use

Idle is what sits in context while the skill is installed. In use is its instructions, loaded when it runs. Counts are approximate, made offline with cl100k_base.

Model dependence Model-led

Script-ledMixedModel-led

Mostly instructions, so how well it works depends on the model that follows them.

This describes the skill's structure, how much of it is scripts. It does not say how well it works.

Touches

Nothing detected: no shell, network, file writes, git or credentials in its files.

Declared tags come from the skill's frontmatter; detected ones from its scripts and code blocks. This is detected, not audited.

How it scores

Quality

How well the skill is written, counted from its own files: its frontmatter, its steps, examples, stated output and constraints, its size and whether every file it ships is one it names.How we score it →
Needs attention
  • Frontmatter8/15
  • Structure0/25
  • Examples0/15
  • Output0/15
  • Size11/15
  • Resources0/15

Security

Starts at 100 and loses points for each thing in the skill's files that a reader should look at first.How we score it →
Excellent

No findings.

Repository health

Pre-computed score based on development activity, maintenance, community, maturity, and trend momentum.How we score it →
Excellent
  • Development activity96
  • Maintenance100
  • Community56
  • Maturity32
  • Momentum40

Grill With Docs is a thin wrapper: its whole body tells the model to call the Skill tool for grilling and for domain-modeling. The README describes the result as a relentless interview about a plan or design that also builds the project’s shared language in GLOSSARY.md and records hard-to-explain decisions as ADRs.

You invoke it yourself; automatic model invocation is disabled.

What You Get

  • An interview that asks detailed questions until the plan or design is resolved
  • A GLOSSARY.md updated inline as terms are agreed
  • ADRs for decisions that are hard to reverse, surprising and the product of a trade-off
  • A shared language that, according to the README, reduces verbosity and keeps variable, function and file names consistent

Common Use Cases

  • Starting any change in a repo where you want the agent aligned with you first
  • Building a shared vocabulary between a team and its agents
  • Preparing the conversation that later becomes a spec with to-spec

What Makes It Unique

It combines two skills by reference, so the interview logic lives in grilling, the glossary logic lives in domain-modeling, and other skills reuse each of them. The README says the shared-language technique might be the single coolest in the repository.

Under The Hood

The SKILL.md holds one instruction: call the Skill tool twice, once for grilling and once for domain-modeling. The behaviour comes from those two skills, which are separate records. It runs no scripts; through domain-modeling it creates and edits GLOSSARY.md and ADR files in your repository.

Join founders building with open source

Opinionated takes, migration guides, cost-saving tips, and insights from the open source ecosystem.

Subscribe on SubstackJoin 750+ subscribers