CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

grill-me

Interview the user relentlessly about a plan or design until reaching shared understanding, resolving each branch of the decision tree. Use when user wants to stress-test a plan, get grilled on their design, or mentions "grill me".

Quality

86%

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

88%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 lean, highly actionable instruction-only skill whose entire body is concrete behavioral guidance with no padding; the interview loop is unambiguous with an explicit stop condition and dependency-ordered questioning. The only flaws are the absence of example questions covering common cases and one dangling license-file reference.

DimensionReasoningScore

Conciseness

The body is roughly 70 words of pure behavioral instruction ("Interview me relentlessly... Walk down each branch... Ask the questions one at a time... explore the codebase instead") with zero concept explanations or padding; every sentence adds a distinct rule. The brief italic footer is attribution plus a useful routing hint to /grill-with-docs, so it earns its place, matching anchor 5.

5 / 5

Actionability

Concrete, executable behavioral rules for an instruction-only skill: "Ask the questions one at a time", "For each question, provide your recommended answer", and "If a question can be answered by exploring the codebase, explore the codebase instead". It stops short of anchor 5 because there are no specific examples covering common cases, such as a sample probing question or the expected dialogue format.

4 / 5

Workflow Clarity

This is a simple single-task skill under 50 lines whose single action (a relentless interview) is unambiguous, with an explicit termination condition ("until we reach a shared understanding") and ordering constraints ("resolving dependencies between decisions one-by-one", one question at a time); the rubric's simple-skill exception applies. No destructive or batch operations are involved, so the validation cap does not apply.

5 / 5

Progressive Disclosure

A single short, well-organized file with no need for external content references, which the rubric notes can score 5 for skills this size. However the one referenced path in the body, ".claude/skills/LICENSE-mattpocock-skills.md", does not exist in this bundle, leaving a minor dangling reference that fits anchor 4 (references mostly clear, minor organization gaps).

4 / 5

Total

18

/

20

Passed

Description

78%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 well-formed description that clearly states what the skill does in third person and provides explicit, natural 'Use when' triggers including the exact phrase a user would say ("grill me"). The main gaps are limited concrete-action coverage (two actions) and a few missing synonyms for the trigger scenario such as "poke holes" or "pressure-test".

Suggestions

Add one or two more concrete actions to the capability sentence (e.g., "asks probing questions one at a time, each with a recommended answer") so specificity moves from two actions to several.

Broaden the trigger clause with common synonyms users actually say for this scenario, such as "poke holes in my plan", "pressure-test my design", or "challenge my assumptions".

DimensionReasoningScore

Specificity

"Interview the user relentlessly about a plan or design" and "resolving each branch of the decision tree" name the domain plus two concrete actions, matching the anchor for 1-2 concrete actions without comprehensive coverage. It falls short of anchor 4 because "until reaching shared understanding" is a goal state rather than an additional specific action, so only two concrete actions are listed.

3 / 5

Completeness

It explicitly answers both what ("Interview the user relentlessly about a plan or design until reaching shared understanding, resolving each branch of the decision tree") and when ("Use when user wants to stress-test a plan, get grilled on their design, or mentions 'grill me'") with concrete trigger phrases, exactly matching anchor 5. Voice is correctly third person ("Interview the user"), so no voice penalty applies.

5 / 5

Trigger Term Quality

"Use when user wants to stress-test a plan, get grilled on their design, or mentions 'grill me'" covers good natural phrases including the exact user utterance "grill me". Common synonyms such as "poke holes in my plan", "pressure-test my design", or "challenge my assumptions" are missing, which fits anchor 4 (a few natural terms missing) rather than the comprehensive anchor 5.

4 / 5

Distinctiveness Conflict Risk

The niche (adversarial interrogation of the user about their plan) is mostly distinct, and "grill me" is a highly distinctive exact-match trigger. However "stress-test a plan" carries minor overlap risk with generic plan-review or critique skills, fitting anchor 4 (mostly distinct, minor overlap with closely related skills) rather than anchor 5's minimal conflict risk.

4 / 5

Total

16

/

20

Passed

Validation

100%

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

agranado2k/google-books-clojure

Path

.claude/skills/grill-me/SKILL.md

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.