CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

ljg-think

追本之箭——纵向深钻思维工具。给一个观点、现象或问题,像箭一样一路向下钻到不可再分的本质。Use when user says '想透', '追本', '本质是什么', '为什么会这样', '深挖', '钻到底', 'think deep', 'drill down', or wants to trace any idea/phenomenon vertically to its irreducible root. Also trigger when user provides a statement and wants depth analysis, not breadth survey.

Quality

82%

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

78%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 is a tight, well-structured instruction skill: explicit drilling rules, a concrete termination test, specific per-layer writing requirements, and copy-paste output commands. The remaining gaps are the absence of a worked example demonstrating the descent format and minor executable details in the output step (topic slugification, directory creation).

Suggestions

Add one short worked example (a sample topic drilled three layers, showing the layer name and the crack leading to the next layer) so the '怎么写' requirements are demonstrated, not just described.

Specify how to form the {主题} filename slug, and add a mkdir -p (or existence check) for ~/Documents/notes/ before writing the file so the output step is fully executable.

DimensionReasoningScore

Conciseness

The ~60-line body is efficient and assumes Claude's competence — no known concepts are explained, and rules are stated tersely ('纵向,不横向——每次下钻必须回答"为什么是这样",不是"还有什么"'). Held at 4 rather than 5 because a few lines are atmospheric rather than instructive: the arrow persona ('一支离弦之箭。射出后只知前进,不知后退') and the dual analogies ('像地质学家追踪地层……像物理学家追问粒子') could be trimmed without losing guidance.

4 / 5

Actionability

Mostly executable guidance: three defined iron rules, explicit layer-count guidance ('浅的话题三层见底,深的七层。你自己判断'), per-layer requirements ('每层有命名——给每一层一个精准的名字,两三个字', '层间有裂缝'), a concrete termination test ('再往下问"为什么",答案要么是同义反复,要么指向上述四类之一'), and copy-paste output commands ('date +%Y%m%dT%H%M%S', the file path pattern). Not level 5: no worked example demonstrating the layer-naming/descent format, and the {主题} filename slug and creation of ~/Documents/notes/ are unspecified.

4 / 5

Workflow Clarity

Clear sequence from drilling logic → method → termination criteria → writing requirements → numbered output steps, with an explicit termination checkpoint ('到底的标志:再往下问"为什么"……'). Not level 5: no validation that the output directory exists or that the note file was written successfully before reporting the path — minor validation gaps consistent with the level-4 anchor. This is not a destructive or batch operation, so no level-3 cap applies.

4 / 5

Progressive Disclosure

A single-purpose skill of ~60 lines with no bundle files (no references/, scripts/, or assets/ exist) and no need for external references; everything is appropriately inlined. Sections are well-organized with distinct roles (你是什么 / 钻的逻辑 / 怎么钻 / 什么算到底 / 怎么写 / 输出), matching the simple-skill exception for a well-organized, self-contained body.

5 / 5

Total

17

/

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.

The description clearly and explicitly states both what the skill does and when to use it, backed by a strong bilingual set of natural trigger phrases. Its weaknesses are a single metaphorical capability statement rather than enumerated concrete actions, and a couple of overly broad triggers that create minor misfire risk in general-analysis contexts.

Suggestions

Replace or supplement the arrow metaphor with 1-2 concrete capability statements (e.g., '对给定观点逐层追问为什么,输出抵达不可再分本质的分层分析') so the 'what' reads as specific actions rather than imagery.

Add missing high-frequency synonyms to the trigger list: '第一性原理 / first principles', '追根溯源', 'root cause / 根本原因'.

Qualify the broadest triggers ('为什么会这样', '深挖') with context (e.g., about ideas/phenomena rather than code errors) so the skill does not fire during debugging conversations.

DimensionReasoningScore

Specificity

Names the domain ('纵向深钻思维工具') and one core action with explicit input and output ('给一个观点、现象或问题,像箭一样一路向下钻到不可再分的本质'), matching the anchor for 1-2 concrete actions without comprehensive coverage. It is more specific than the generic level 2, but does not list several distinct actions needed for level 4 — the capability is stated once, metaphorically.

3 / 5

Completeness

Explicitly answers both questions: what ('给一个观点、现象或问题……一路向下钻到不可再分的本质') and when ('Use when user says [8 concrete phrases]… Also trigger when user provides a statement and wants depth analysis, not breadth survey'). This matches the level-5 anchor of clear what AND when with concrete trigger phrases, not merely the weaker 'when' of level 4.

5 / 5

Trigger Term Quality

Eight natural trigger phrases across both languages ('想透', '追本', '本质是什么', '为什么会这样', '深挖', '钻到底', 'think deep', 'drill down') plus two situational conditions ('wants to trace any idea/phenomenon vertically to its irreducible root', 'wants depth analysis, not breadth survey') — good, natural coverage. Held at 4 rather than 5 because common variants users would actually say are missing: '第一性原理/first principles', '追根溯源', 'root cause/根本原因', 'dig/go deeper'.

4 / 5

Distinctiveness Conflict Risk

A clear niche (vertical essence-tracing, explicitly 'depth analysis, not breadth survey') with dedicated triggers like '追本' and '钻到底' makes it mostly distinct. Not level 5: broad phrases such as '为什么会这样' and '深挖' would also naturally occur in debugging or general root-cause conversations, creating minor overlap risk with general analysis skills; not level 3 because the framing and dedicated vocabulary clearly separate it from generic analysis.

4 / 5

Total

16

/

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-think/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.