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

88%

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

92%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.

An exemplary lean, instruction-only skill body: every sentence is an executable directive, the process has an unambiguous sequence and end state, and it needs no progressive disclosure at this size. The only gap is minor — no example question and no handling of the case where the user has not yet shared a plan.

Suggestions

Add a brief note on what to do when no plan or design is on the table (e.g., ask the user to share it first) to close the one behavioral gap.

Optionally include one example question showing the expected format (question + recommended answer) to make the interaction pattern concrete.

Dimension	Reasoning	Score
Conciseness	Three lean paragraphs with zero padding and no explanation of concepts Claude already knows; every instruction earns its place, matching the 'lean and efficient; assumes Claude's competence' anchor.	5 / 5
Actionability	For an instruction-only skill the directives are concrete and directly executable ('Ask the questions one at a time', 'For each question, provide your recommended answer', 'If a question can be answered by exploring the codebase, explore the codebase instead'). Not a 5 because it includes no example question format and does not address what to do when no plan or design has been provided.	4 / 5
Workflow Clarity	A simple, single-purpose skill whose process is unambiguous: questions one at a time, walk each branch of the design tree resolving dependencies, with a defined end state ('until we reach a shared understanding'). Per the simple-skill exception this earns 5; no destructive or batch operations require validation checkpoints.	5 / 5
Progressive Disclosure	The body is under 50 lines, self-contained, requires no external references, and no bundle files exist, so nothing is inlined that belongs in a separate file. Per the under-50-lines guideline, the flat structure fully earns a 5.	5 / 5
	Total	19 / 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 explicitly states both what the skill does and when to use it, with natural trigger phrases including the literal 'grill me' invocation. Its main limitation is specificity: it describes one core action rather than enumerating several discrete capabilities, which is partly inherent to a single-purpose skill.

Suggestions

Enumerate the concrete behaviors more discretely in the what-clause (e.g., 'asks probing questions one at a time, recommends an answer for each, and resolves dependencies between decisions') to raise specificity.

Add one or two more natural trigger variations users might say (e.g., 'poke holes in my plan', 'interrogate my design') to broaden trigger coverage without adding fluff.

Dimension	Reasoning	Score
Specificity	Names the domain and one-two concrete actions ('Interview the user relentlessly', 'resolving each branch of the decision tree'), but does not enumerate several discrete actions as the 4/5 anchors require. Not a 4 because the description describes a single action two ways rather than listing multiple specific operations.	3 / 5
Completeness	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, matching the anchor-5 example pattern exactly.	5 / 5
Trigger Term Quality	Includes natural phrases users would actually say: 'stress-test a plan', 'get grilled on their design', and the literal invocation 'grill me'. Not a 5 because coverage of synonym space (e.g., 'interrogate me', 'poke holes in my design', 'challenge my plan') is incomplete.	4 / 5
Distinctiveness Conflict Risk	Clear niche (adversarial interview about a plan/design) with a distinctive trigger phrase ('grill me') and minimal conflict risk. Not a 5 because 'stress-test a plan' has minor overlap risk with general plan-review or design-review skills.	4 / 5
	Total	16 / 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: webiny/webiny-js
Path: .claude/skills/grill-me/SKILL.md
Commit: f0d3d3e

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.