CtrlK
BlogDocsLog inGet started
Tessl Logo

grill-with-docs

Grilling session that challenges your plan against the existing domain model, sharpens terminology, and updates documentation (CONTEXT.md, ADRs) inline as decisions crystallise. Use when the user wants to stress-test a plan against their project's language and documented decisions, or asks to "grill me with docs" or "interview me about this".

74

Quality

92%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

SecuritybySnyk

Passed

No findings from the security scan

SKILL.md
Quality
Evals
Security

Quality

Content

90%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A well-crafted instruction skill body: concise, assumes Claude's competence, and packed with concrete directives, a decision checklist, and model phrasings. The main gap is that the two referenced format files (CONTEXT-FORMAT.md, ADR-FORMAT.md) are not bundled, leaving dead links where detailed formatting guidance is expected.

Suggestions

Bundle the referenced CONTEXT-FORMAT.md and ADR-FORMAT.md (or inline their formats) so the links in 'Update CONTEXT.md inline' and 'Offer ADRs sparingly' resolve to real content.

Add an explicit verification checkpoint — e.g. confirm a resolved term is recorded in CONTEXT.md before moving to the next question — to strengthen the workflow's feedback loop.

Consider tightening the two ASCII file-structure trees into one combined example to reduce tokens while preserving the single-vs-multi-context distinction.

DimensionReasoningScore

Conciseness

Lean and efficient throughout — it references a 'Nygard-style ADR' and ADRs without explaining what they are, gives terse directives ('Create files lazily, only when you have something to write'), and every section including the ASCII structure trees earns its place.

5 / 5

Actionability

Highly actionable for an instruction skill: concrete directives ('Ask the questions one at a time, waiting for feedback'), a crisp 3-condition ADR checklist, named format references, and copy-paste-ready example phrasings ('Your glossary defines "cancellation" as X, but you seem to mean Y').

5 / 5

Workflow Clarity

Clear sequenced discipline — grill one question at a time, propose a recommended answer, wait for feedback, explore the codebase when possible, update docs inline — with a built-in feedback loop and the 3-condition ADR checklist, but no explicit validate-then-proceed gates of the kind the top anchor describes.

4 / 5

Progressive Disclosure

Well-sectioned body (Domain awareness, During the session, Pairs With) with one-level-deep, clearly signaled references to CONTEXT-FORMAT.md and ADR-FORMAT.md, but neither referenced file is present in the bundle, so navigation dead-ends.

4 / 5

Total

18

/

20

Passed

Description

95%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A strong description that uses third-person voice, names concrete actions with specific artifacts (CONTEXT.md, ADRs), and provides an explicit 'Use when' clause with quoted natural trigger phrases. Its only weakness is that the lead action ('challenges your plan against the existing domain model') is slightly abstract compared to the crisper actions that follow.

DimensionReasoningScore

Specificity

Lists several concrete actions — 'sharpens terminology' and 'updates documentation (CONTEXT.md, ADRs) inline' — but 'challenges your plan against the existing domain model' is more abstract framing than a crisp discrete action, leaving a minor coverage gap.

4 / 5

Completeness

Explicitly answers both what ('Grilling session that challenges... sharpens terminology, and updates documentation... inline') and when ('Use when the user wants to stress-test a plan... or asks to grill me with docs...') with concrete trigger phrases.

5 / 5

Trigger Term Quality

Comprehensive natural trigger coverage with quoted user phrases users would actually say — 'grill me with docs', 'interview me about this' — plus the synonym 'stress-test a plan against their project's language and documented decisions'.

5 / 5

Distinctiveness Conflict Risk

Clear niche — interviewing the user against an existing domain model and documentation — with distinct triggers ('grill me with docs', 'interview me about this') that minimize overlap with other skills.

5 / 5

Total

19

/

20

Passed

Validation

93%

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

relative_links

Relative link issues: 2 missing

Warning

Total

15

/

16

Passed

Repository
tslateman/duet
Reviewed

Table of Contents

Is this your skill?

If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.