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

81%

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

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

Quality

Content

72%Scale 1-3

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

This is a concise, well-scoped conversational skill that efficiently communicates its core intent. Its main weakness is a lack of concrete examples (e.g., sample questions, a sample decision-tree branch, or criteria for 'resolved') and missing explicit completion/validation criteria, which would make the guidance more actionable and the workflow clearer.

Suggestions

Add 1-2 concrete example exchanges showing what a good probing question and resolution looks like, so Claude has a clear quality bar.

Define explicit criteria for when a branch of the decision tree is considered 'resolved' and when the overall interview is complete.

Specify what 'provide your recommended answer' should look like—e.g., a brief rationale with tradeoffs, a single-sentence recommendation, etc.

Dimension	Reasoning	Score
Conciseness	The skill is extremely lean—just four short directives with no unnecessary explanation. Every sentence adds a distinct behavioral instruction. Claude already understands interviewing and codebase exploration, so no concept explanation is needed.	3 / 3
Actionability	The instructions are clear behavioral directives but remain somewhat abstract—there are no concrete examples of what a good question looks like, no example dialogue, and no specific criteria for when a branch is 'resolved.' It tells Claude what to do but not how to judge quality.	2 / 3
Workflow Clarity	There is an implicit workflow (ask questions one at a time, walk decision tree branches, resolve dependencies), but the sequence lacks explicit checkpoints—e.g., how to determine when a branch is fully resolved, when to move on, or when the overall interview is complete.	2 / 3
Progressive Disclosure	For a simple, single-purpose skill under 50 lines with no need for external references, the content is appropriately organized in a single concise block. No external files are needed.	3 / 3
	Total	10 / 12 Passed

Description

89%Scale 1-3

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

This is a well-crafted description with clear 'what' and 'when' clauses, distinctive trigger terms like 'grill me' and 'stress-test', and a unique niche that minimizes conflict risk. The main weakness is that the specific actions could be more granular — listing concrete techniques like probing assumptions, identifying edge cases, or challenging trade-offs would strengthen specificity.

Suggestions

Add more specific concrete actions, e.g., 'probes assumptions, identifies edge cases, challenges trade-offs, and surfaces hidden dependencies' to strengthen specificity.

Dimension	Reasoning	Score
Specificity	The description names the domain (plan/design interrogation) and describes the core action ('interview relentlessly', 'resolving each branch of the decision tree'), but doesn't list multiple concrete sub-actions beyond the general interviewing process.	2 / 3
Completeness	Clearly 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"').	3 / 3
Trigger Term Quality	Includes strong natural trigger terms: 'stress-test a plan', 'get grilled on their design', 'grill me', 'plan', 'design'. These are phrases users would naturally say when seeking this kind of interaction.	3 / 3
Distinctiveness Conflict Risk	This skill has a very distinct niche — adversarial/Socratic interrogation of plans and designs. The trigger 'grill me' is highly distinctive, and the confrontational interview style clearly separates it from general planning or design review skills.	3 / 3
	Total	11 / 12 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 — 11 / 11 Passed

Validation for skill structure

No warnings or errors.

Repository: SnowingFox/ai-skills
Path: skills/grill-me/SKILL.md
Commit: d6882ad

Reviewed: ~13 hours 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.