CtrlK
BlogDocsLog inGet started
Tessl Logo

ljg-constraint

给一个领域、专业、角色、产品或争论找出真正框住它的几条约束,判明它们属于世界/规则/解释(硬/软/自设),看这组约束如何定义身份、补全问题、框出解空间并解释实际行为;尤其用于区分目标相同但约束不同导致的方案分歧,识别被误当硬事实的旧解释。USE WHEN 用户说 '约束', '找约束', '这个领域的约束是什么', '本质约束', '什么框住了它', '硬约束还是自设约束', '边界在哪', '这行为什么这么做', 'constraint', '/ljg-constraint', or gives any domain/profession/role/product/debate and wants its defining constraints and solution space mapped. NOT FOR 往下挖生成器(用 ljg-rank,找的是生成力不是边界)、纵向深钻一个观点到底(用 ljg-think)、拆一本书(用 ljg-book)、一个概念解剖(用 ljg-learn)。

69

Quality

87%

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

70%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 methodologically strong body with an unusually good validation loop (心内判据 checklist with retry) and concrete templates, examples, and output commands. Its weaknesses are redundancy (the description-over-modification theme repeated 4+ times) and a monolithic ~240-line single-file structure where the diagram specs and style guide could live in references/.

Suggestions

State the 'describe first, don't rush to solutions' rule once (e.g., in a single prominent place) and cut its repetitions from the Gotchas, 约束是什么, 真伪定性, and 松动哪条 sections — the repetition costs several hundred tokens without adding guidance.

Move the three ASCII diagram specifications (and optionally the 中文母语化 table) into a references/ file (e.g. references/diagrams.md), keeping SKILL.md a lean overview that names the three shapes and links out — mirroring how template.org is already externalized.

Add a short excerpt of a finished output essay (or one fully worked constraint->solution-space->behavior chain in prose form) so the writing style being demanded is demonstrated, not just described.

DimensionReasoningScore

Conciseness

The 'description before modification / don't rush to find exits' theme is repeated at least four times (Gotchas "不要急着写'如何突破约束'", "这引擎首先是描述,不是改造", "注意这一步是在更准确地描述...不是在指路", and the entire "松动哪条(第二位)" section), and "约束是什么" opens with philosophical prose (water/cup metaphor) that could be tightened. Not a 4 ('minor instances') because the redundancy is sectional, not minor; not a 2 because the bulk — hardness layers, discriminating questions, diagram specs, judging gates — is genuinely non-obvious methodology Claude does not already know, not padding.

3 / 5

Actionability

Concrete guidance is strong: per-constraint discriminating questions ("违背它,是系统物理崩溃(硬),还是有人罚你(软)"), copy-paste output commands (`date +%Y%m%dT%H%M%S`, `~/Documents/notes/{时间戳}--{领域}的约束__constraint.org` with frontmatter in the real file `references/template.org`), three complete ASCII diagram templates, and three worked input->output examples. Not a 5 because it never shows even a fragment of the finished essay it demands, and the writing-quality criteria ("一口气读完", "有落差") remain stylistic direction rather than executable checks.

4 / 5

Workflow Clarity

Clear sequenced workflow with explicit validation and feedback loops: a Workflow Routing table maps each trigger to a workflow; the 7-step 工序 is explicitly ordered; the 心内判据 section is a 7-item checklist with a retry loop ("过不了一条,回工序接着走") and 解释力 as the primary gate; the diagram section adds "右边对不上左边,就是约束没找准,回去接着挖". This matches the 5 anchor (clear sequence, explicit validation, feedback loops, checklist).

5 / 5

Progressive Disclosure

The one bundle reference (`references/template.org` via "frontmatter 见 references/template.org") is real, clearly signaled, and one level deep — but the 240-line / ~20KB body is largely monolithic: the three full diagram specifications with examples, the Chinese-nativization table, and the essay-writing guide are inlined material that plausibly belongs in reference files, matching the 3 anchor ('content that should be separate is inline'). Not a 2 because section headers make it navigable and the template IS appropriately externalized; not a 4 because the inline bulk goes well beyond a lean overview plus well-signaled references.

3 / 5

Total

15

/

20

Passed

Description

100%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.

An exemplary description: concrete capability list, explicit USE WHEN triggers with natural user phrasings including the slash command, and an explicit NOT FOR boundary against sibling skills. It is long but every clause is functional (what / when / not-for) rather than padded.

DimensionReasoningScore

Specificity

The description lists multiple concrete actions: "找出真正框住它的几条约束", "判明它们属于世界/规则/解释(硬/软/自设)", "框出解空间并解释实际行为", and "识别被误当硬事实的旧解释" — comprehensive coverage of the skill's capabilities with no material gaps, matching the 5 anchor rather than the 4 ('minor gaps in coverage').

5 / 5

Completeness

It explicitly answers both 'what' (find and classify constraints into hard/soft/self-imposed layers, map the solution space, explain behavior) and 'when' via a concrete "USE WHEN ... 用户说 ..." clause, plus a "NOT FOR" exclusion set. This is the 5 anchor: both what and when with concrete trigger phrases.

5 / 5

Trigger Term Quality

It covers natural user phrasings ('约束', '找约束', '本质约束', '什么框住了它', '硬约束还是自设约束', '边界在哪', '这行为什么这么做') plus the English term 'constraint' and the slash command '/ljg-constraint', and a general trigger for any domain/role/debate input. This matches the 5 anchor (synonyms plus invocation name); the 4 anchor requires 'a few natural terms missing', which is not the case.

5 / 5

Distinctiveness Conflict Risk

The "NOT FOR" clause explicitly redirects sibling skills (ljg-rank "找的是生成力不是边界", ljg-think, ljg-book, ljg-learn) and the niche (constraint/boundary analysis vs. generator digging) is clearly distinct with its own trigger terms, matching the 5 anchor (clear niche, distinct triggers, minimal conflict risk).

5 / 5

Total

20

/

20

Passed

Validation

93%

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

Validation — 15 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

frontmatter_unknown_keys

Unknown frontmatter key(s) found; consider removing or moving to metadata

Warning

Total

15

/

16

Passed

Repository
lijigang/ljg-skills
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.