CtrlK
BlogDocsLog inGet started
Tessl Logo

ljg-mechanism

针对用户提供的现象,用「看反常—追过程—分解释—带条件迁移」生成机制猜想,并联网核对现象、检验关键因果与竞争解释,交付证据最能支撑的机制及其边界。USE WHEN 用户调用 ljg-mechanism、说「找机制」「机制研究」「用四步框架走一遍」「联网验证这个解释」,或希望研究一个现象怎样发生。NOT FOR 纯概念定义、普通摘要、已有唯一解的操作题或只画地图。

64

Quality

80%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide
SecuritybySnyk

Low

Low-risk findings worth noting

Fix and improve this skill with Tessl

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

Quality

Content

61%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 lean, well-organized instruction body with unusually strong epistemic guardrails and good worked examples, undermined by a dangling primary reference: Workflows/DiscoverMechanism.md is missing, so the core four-step procedure is never actually defined anywhere in the bundle. Actionability and workflow clarity suffer accordingly.

Suggestions

Add the missing Workflows/DiscoverMechanism.md (or inline the four-step procedure in the body) so the routing table's primary entry point resolves and the steps 看反常/追过程/分解释/带条件迁移 are operationally defined.

For each of the four steps, specify one or two concrete moves or checks (e.g. what counts as an anomaly worth chasing, how to phrase a distinguishing prediction between competing explanations) so the body is actionable even without the workflow file.

If no external detail is needed, drop the Workflows reference entirely and rely on the well-organized in-body sections, which alone would satisfy progressive disclosure for a sub-50-line skill.

DimensionReasoningScore

Conciseness

The ~36-line body is dense and lean with zero padding: it explains nothing Claude already knows, every bullet carries a non-obvious guardrail (e.g. 「不给每个简单问题硬加图、公式或固定数量的假说」, 「把同一论文的多次转述计作多个独立证据,会夸大支持」). Every token earns its place, matching the 5 anchor.

5 / 5

Actionability

There is concrete guidance (mechanism statements must specify 「谁或什么,在什么条件下,通过哪些中间变化,生成什么结果」; sources placed next to judgments with evidence-type labels; offline mode 「不能声称已联网验证」; three worked examples), but the core procedure for executing the four steps is delegated to Workflows/DiscoverMechanism.md, which does not exist in the bundle. Key details are missing, fitting the 3 anchor rather than 4, where gaps would be minor.

3 / 5

Workflow Clarity

The routing table sequences the main cases and 研究约定 supplies validation checkpoints (verify the raw phenomenon, test competing explanations, distinguish evidence scope), but the four-step sequence 「看反常—追过程—分解释—带条件迁移」 is only ever named, never elaborated, and the file that would contain the workflow is absent. Steps are listed with real gaps, matching the 3 anchor.

3 / 5

Progressive Disclosure

The body itself is well-sectioned and the one reference is clearly signaled in the routing table, but Workflows/DiscoverMechanism.md — the primary navigation target and the only reference in the skill — does not exist in the bundle (no references/, scripts/, assets/, or Workflows/ directories are present). A clearly signaled yet nonexistent main reference is a hard navigation failure, below the 3 anchor where references are present but weakly signaled.

2 / 5

Total

13

/

20

Passed

Description

92%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: concrete capabilities in third person, explicit USE WHEN with natural quoted trigger phrases, a NOT FOR boundary clause, and a distinct niche. The only weakness is a few missing natural trigger variations.

DimensionReasoningScore

Specificity

The description lists multiple concrete actions — 「用『看反常—追过程—分解释—带条件迁移』生成机制猜想」、「联网核对现象」、「检验关键因果与竞争解释」、「交付证据最能支撑的机制及其边界」 — comprehensively covering generation, verification, testing, and delivery for this niche. There is no coverage gap, so it matches the 5 anchor rather than 4.

5 / 5

Completeness

It clearly answers both what (four-step mechanism-hypothesis generation, online verification, causal and competing-explanation testing, delivery with boundaries) and when (explicit 「USE WHEN」 with concrete trigger phrases, plus a 「NOT FOR」 clause). This matches the 5 anchor exactly.

5 / 5

Trigger Term Quality

Explicit natural quoted phrases (「找机制」「机制研究」「用四步框架走一遍」「联网验证这个解释」) plus the skill-name invocation give good keyword coverage, but common user variations like 「为什么会这样」/「原因分析」 are missing. Fits the 4 anchor (good coverage, a few natural terms missing), clearly above 3 (several natural phrases are present).

4 / 5

Distinctiveness Conflict Risk

A clear niche (mechanism/causal research on phenomena) with distinct quoted triggers, and the 「NOT FOR 纯概念定义、普通摘要、已有唯一解的操作题或只画地图」 clause explicitly fences off adjacent skills. Minimal conflict risk.

5 / 5

Total

19

/

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

relative_links

Relative link issues: 1 missing

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.