grill-with-docs

Relentless interview that also writes ADRs and glossary. Trigger: /grill-with-docs.

Quality

74%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./ywai/skills/grill-with-docs/SKILL.md

The canonical home for this skill is grill-with-docs in mattpocock/skills

Quality

Content

82%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.

The body is a maximally lean dispatcher: one unambiguous sentence delegating to two named skills. It fully respects the token budget and carries no known-concept padding; its only weakness is the absence of any ordering rationale, argument passing, or post-invocation guidance.

Dimension	Reasoning	Score
Conciseness	The body is a single 12-word sentence: "Call the Skill tool twice, for "grilling" and "domain-modeling"." Every token carries instruction weight, explains nothing Claude already knows, and assumes competence. This is the lean anchor 5.	5 / 5
Actionability	The instruction is executable as written — two Skill invocations with the named skills "grilling" and "domain-modeling" — which is concrete, not vague. It stops short of anchor 5 because no arguments, outputs, or per-call context are specified, leaving minor gaps for the model to fill.	4 / 5
Workflow Clarity	As a simple dispatcher skill (<50 lines, no destructive or batch operations), the simple-skill exception applies and no validation checkpoints are required. "twice, for 'grilling' and 'domain-modeling'" establishes an unambiguous order, but whether grilling output feeds domain-modeling and what to do after the second call are unstated, keeping it below 5.	4 / 5
Progressive Disclosure	There are no bundle files (no references/, scripts/, assets/) and nothing that warrants offloading, so one-level-deep references are not needed. However the body is a bare line with no section organization at all, which falls just short of the 'well-organized sections' bar for a 5 on even a simple skill.	4 / 5
	Total	17 / 20 Passed

Description

53%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.

The description names a distinct niche and two concrete artifacts, but reads more like a label than a trigger description: no natural-language 'use when' guidance, sparse synonyms, and a vague sense of what the skill actually does. Its slash-command trigger partially compensates for the missing usage clause.

Suggestions

Replace the slash-command trigger line with a natural-language usage clause, e.g. 'Use when the user wants to be grilled on a design or codebase, wants a written record of decisions, or asks for a glossary/ADR.'

Expand trigger vocabulary with the phrases users actually say: 'grill me', 'quiz me', 'interview me about this design', 'write ADRs', 'architecture decision records', 'build a glossary'.

State the concrete actions performed: e.g. 'Asks increasingly hard probing questions, then records accepted decisions as ADRs and compiles a domain glossary.'

Dimension	Reasoning	Score
Specificity	"Relentless interview that also writes ADRs and glossary" names the domain (interviewing) and two concrete outputs (ADRs, glossary), but coverage is thin and "relentless" adds no capability information. It sits between anchor 2 (domain named, actions generic) and anchor 4 (several specific actions); it names real artifacts but only 1-2 actions.	3 / 5
Completeness	The 'what' is present but loose ("relentless interview that also writes ADRs and glossary" leaves the actual behavior undefined), and the 'when' reduces to the slash command "/grill-with-docs" with no conditions for use. A missing 'Use when...' clause or equivalent usage context caps completeness at 3 per the judging guidelines.	3 / 5
Trigger Term Quality	The only usable keywords are "interview", "ADRs", and "glossary"; natural user phrasings like "grill me", "ask me questions", "quiz me", "architecture decision record", or "document the domain" are absent, and "Trigger: /grill-with-docs" is a command name rather than natural language. This matches anchor 3 (some relevant keywords, missing common variations/synonyms), not 4 which requires good coverage.	3 / 5
Distinctiveness Conflict Risk	"grill-with-docs", ADRs, and glossary writing form a fairly distinct niche unlikely to collide with unrelated skills, though the generic word "interview" could overlap with interviewing/candidate-prep or Q&A skills. Mostly distinct with minor overlap risk matches anchor 4.	4 / 5
	Total	13 / 20 Passed

Validation

100%

Warnings & errors only

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

Validation — 16 / 16 Passed

Validation for skill structure

No warnings or errors.

Repository: YoizenSA/dev-ai-workflow
Path: ywai/skills/grill-with-docs/SKILL.md
Commit: 36f5af5

Reviewed: 3 days ago

Table of Contents

Discovery Implementation Validation

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.