grilling

围绕计划、decision 或 idea 持续追问用户。适用于用户想对自己的思路做压力测试，或使用任何 “grill” 触发措辞时。

Quality

78%

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 ./skills/productivity/grilling/SKILL.md

Quality

Content

86%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 tight, well-designed instruction skill: it defines a precise iterative interview procedure (design tree, frontier, rounds) with a concrete output format, clear rules for what to ask the user versus look up via sub-agents, and an explicit termination-plus-confirmation gate. The only weaknesses are a few unspecified edge cases in the question loop, which cost it the top actionability and workflow-clarity anchors.

Dimension	Reasoning	Score
Conciseness	The body is lean and every sentence earns its place: it introduces the design tree/frontier model, the round format, the fact-vs-decision division of labor, and the termination condition with zero padding or explanation of concepts Claude already knows. This matches the 'lean and efficient; every token earns its place' anchor.	5 / 5
Actionability	Concrete, executable guidance throughout: an exact per-round format template ("❓ Q1 - ... ➡️ <你的推荐答案>"), the rule to number questions and attach recommended answers, and specific directives like "派一个 sub-agent 去查" for environment facts and "逐个交给他，并等待" for decisions. It falls short of 5 because a few execution details are left open, such as how (or whether) to surface the design tree to the user and how to handle answers that contradict an earlier settled decision.	4 / 5
Workflow Clarity	The multi-step process is clearly sequenced with a genuine feedback loop: build the tree, compute the frontier ("prerequisites 已经敲定的 decisions"), ask a full numbered round, wait for answers, then "重新计算 frontier" and repeat. It has an explicit stop condition ("当 frontier 为空时") and a confirmation gate ("在用户确认我们已经达成共同理解之前，不要采取行动"). It does not reach 5 because there is no guidance for recovering from conflicting or ambiguous user answers within the loop.	4 / 5
Progressive Disclosure	This is a single-purpose skill under 50 lines with no need for external references — and the bundle indeed contains no references/, scripts/, or assets/ files, so nothing is missing. The content is clear and logically ordered (model → procedure → format → dynamics → division of labor → termination), which the rubric's simple-skill exception says can score 5.	5 / 5
	Total	18 / 20 Passed

Description

70%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 concise, third-person description that clearly states what the skill does and gives explicit trigger guidance including a distinctive keyword ('grill'). Its main limitations are a single-action 'what' statement and moderate trigger synonym coverage, which leave it just short of the top anchors.

Suggestions

Expand the 'what' with 1-2 more concrete actions or the outcome, e.g. '持续追问用户的计划/决策直到达成共同理解，再开始行动' (persistently interrogate the user's plan/decision until shared understanding is reached before acting).

Add a couple of natural trigger variations users might say, such as '帮我挑战这个方案' / 'poke holes in my plan', alongside the existing 'grill' and '压力测试'.

Consider mirroring key trigger terms in both Chinese and English so the description matches regardless of the language the user types in.

Dimension	Reasoning	Score
Specificity	The description names its domain ("计划、decision 或 idea") and one concrete action ("持续追问用户"), matching the '1-2 concrete actions, not comprehensive' anchor. It stays at 3 rather than 4 because no additional specific actions or outcomes are enumerated.	3 / 5
Completeness	It explicitly answers both 'what' ("围绕计划、decision 或 idea 持续追问用户") and 'when' ("适用于用户想对自己的思路做压力测试，或使用任何 'grill' 触发措辞时") with a concrete trigger phrase. It falls short of 5 because the 'what' is a single brief action and does not convey the skill's outcome (reaching shared understanding before acting).	4 / 5
Trigger Term Quality	It includes natural trigger language users would actually say — the explicit "grill" trigger phrasing and "压力测试" (stress-test) — plus plan/decision/idea vocabulary. A few natural synonyms (e.g. 'poke holes in', 'challenge my plan') are missing, so it does not reach comprehensive coverage.	4 / 5
Distinctiveness Conflict Risk	The niche is distinct — user-directed interrogation/stress-testing of one's own thinking — anchored by the distinctive "grill" trigger term, so wrong-skill triggering is unlikely. Minor overlap remains with generic brainstorming or planning-review skills, keeping it below 5.	4 / 5
	Total	15 / 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: vinvcn/mattpocock-skills-zh-CN
Path: skills/productivity/grilling/SKILL.md
Commit: 3f92a83

Reviewed: 4 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.