CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

ljg-structure

找出一段信息中母题级别的结构,把抽象关系展开回具体可见的现象,再用风洞检验关键因果与边界。USE WHEN user says '找结构', '母题是什么', '结构风洞', '背后的结构', '不要只做AB测试', or wants the generative mechanism beneath a problem. 仅当用户明确要求时展开跨域来源或多母题组合。NOT FOR 普通摘要、单个类比、事实查询或已有明确唯一解的执行题。

Quality

77%

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

Fix and improve this skill with Tessl

tessl review fix ./skills/ljg-structure/SKILL.md
SKILL.md
Quality
Evals
Security

Quality

Content

57%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.

The body delivers unusually concrete output discipline — exact headings, commands, file template, and length budget — supported by well-chosen examples. Its two real defects are the missing 'Workflows/FindStructure.md' file behind the routing table and the absence of an explicit ordered workflow, plus meaningful repetition of the wind-tunnel rules across three sections.

Suggestions

Fix the Workflow Routing table: either ship 'Workflows/FindStructure.md' in the bundle or remove the routing section, since following it currently leads to a missing file.

Add one short numbered sequence (输入 → 找母题 → 结构卡 → 内部风洞 → 结论 → 落盘) early in the body so the pipeline does not have to be reconstructed from the Gotchas and Examples.

De-duplicate the wind-tunnel/conclusion rules: state them once in '风洞如何进入结论' and trim the overlapping Gotchas items (e.g. the repeated '结论不是全文摘要' / '具体试压与抽象规则必须递进' pair).

DimensionReasoningScore

Conciseness

The body is dense and rule-per-line with no padding or explanation of known concepts, but the wind-tunnel-to-conclusion rule is stated three times (理解契约 section, Gotchas items on 风洞/结论, and again in 输出要求) and several Gotchas restate 理解契约 points. Fits 'mostly efficient but could be tightened' rather than the noticeably-verbose 2 anchor.

3 / 5

Actionability

Concrete, executable guidance throughout: exact commands ('date +%Y%m%dT%H%M%S', 'date "+%Y-%m-%d %a %H:%M"'), a copy-paste file path template, a full markdown skeleton with required frontmatter, strict heading and ASCII-connector rules, and three worked examples. Not 5 because the core analytical moves (how to actually locate the 母题, how to lay out the relation graph) remain conceptual.

4 / 5

Workflow Clarity

The pipeline (输入 → 母题 → 结构卡 → 内部风洞 → 结论 → 落盘) is never presented as an explicit ordered sequence; it must be synthesized from 理解契约, Gotchas, 输出要求, and the Examples. One validation checkpoint exists ('保存后读回文件并报告路径'), but the sole workflow reference 'Workflows/FindStructure.md' does not exist in the bundle. Non-destructive skill, so the batch cap does not apply; fits 'sequence present but implicit'.

3 / 5

Progressive Disclosure

The body's only external reference — 'Workflows/FindStructure.md' in the Workflow Routing table — is a dangling pointer (no references/, scripts/, assets/, or Workflows/ directory exists in the bundle). Internal sections are well organized, but a broken sole reference is more than the 'minor organization gaps' of the 4 anchor, landing at 'references present but not clearly (or validly) signaled'.

3 / 5

Total

13

/

20

Passed

Description

87%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 strong description: it names three concrete capabilities, gives explicit USE WHEN triggers in natural user language, and closes with clear exclusions. The only weakness is that its key terms ('母题', '风洞') are the skill's own coinage, so a reader outside the skill's context may need one beat to decode them.

DimensionReasoningScore

Specificity

Lists three distinct actions — '找出一段信息中母题级别的结构', '把抽象关系展开回具体可见的现象', '再用风洞检验关键因果与边界' — which go beyond the 1-2-action anchor. Falls short of 5 because the actions are phrased in the skill's own coined vocabulary ('母题级别', '风洞') rather than fully self-explanatory concrete operations.

4 / 5

Completeness

Explicitly answers both 'what' (find mother-theme structure, expand to concrete phenomena, wind-tunnel test causality) and 'when' with concrete quoted trigger phrases, and adds a NOT FOR exclusion list. Matches the 5 anchor ('clearly and explicitly answers both what AND when with concrete trigger phrases'); the 4 anchor's weaker 'when' does not apply.

5 / 5

Trigger Term Quality

Five natural quoted trigger phrases ('找结构', '母题是什么', '结构风洞', '背后的结构', '不要只做AB测试') plus a semantic trigger ('wants the generative mechanism beneath a problem') — phrases a user would plausibly say verbatim. Not 5 because common variants like '深层机制' or '本质结构' are absent.

4 / 5

Distinctiveness Conflict Risk

Clear niche (generative-mechanism / structural analysis) with distinctive triggers ('结构风洞', '母题') and an explicit exclusion clause ('NOT FOR 普通摘要、单个类比、事实查询或已有明确唯一解的执行题') that prevents overlap with summary, analogy, and lookup skills.

5 / 5

Total

18

/

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

Path

skills/ljg-structure/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.