CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

ljg-blind

盲区扫描——读昨天你与 AI 的全部对话,照出暴露的思维盲区(不是不懂的知识,是让某类真相一直看不见的思维习惯),再从微信读书挑一本书的一章精准补上,落成一篇完整分析笔记。Use when user says '扫盲区', '盲区', '照盲区', '看看我的思维盲区', '我昨天想漏了什么', 'blind spot', 'ljg-blind', '/ljg-blind', or wants yesterday's AI conversations analyzed for cognitive blind spots plus a targeted weread chapter to fill them. 可选参数:传一个日期(YYYY-MM-DD)扫那天而非昨天。NOT FOR 纵向深钻一个观点(用 ljg-think)、一个领域降秩(用 ljg-rank)、陪读一篇文本(用 ljg-read)、找一个领域的约束(用 ljg-constraint)。

Quality

94%

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

88%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 strong, highly operational skill body: every step is executable with copy-paste commands, the workflow has real validation checkpoints and feedback loops, and failure modes are anticipated via 兜底 paths and a Gotchas section. The only deductions are minor — a bit of markdown-basic over-explanation, some inlinable style content, and a small '三步/四步' inconsistency.

DimensionReasoningScore

Conciseness

Mostly tight and functional — the five signals, selection criteria, and step-by-step commands are dense with no padding — but the 'Markdown 严格语法' section explains markdown basics Claude already knows (and repeats the heading rule twice), and '和「早信」的分工' is informational context rather than execution guidance. These are minor trimmable instances, so it sits between the efficient (4) and every-token-earns-its-place (5) anchors.

4 / 5

Actionability

Fully executable throughout: complete bash with jq pipelines handling UTC-vs-local-date filtering and subagent exclusion, copy-paste curl payloads for search and chapter lookup, a concrete reading-time formula ('wordCount / 280 向上圆到 5 的倍数'), an exact output path pattern, and a full note template. Placeholders like <核心词> and <bookId> only mark genuinely variable inputs, which the common cases anchor permits.

5 / 5

Workflow Clarity

Five clearly sequenced steps with explicit validation and feedback loops: the data-thinness checkpoint ('当天真人输入 <200 字…不强造盲区,宁可交白卷'), a 7-item 自检 checklist run before finishing with an explicit loop back ('没有就退回第三步'), a search-failure 兜底, and a Gotchas section covering known failure modes. This matches the top anchor's checklist + error-recovery pattern.

5 / 5

Progressive Disclosure

Good structure with clear sections, and the single external tool reference (Tools/WeReadWebUrl.ts) is well signaled with its expected output shown. It falls short of a 5 because no content is split out at all — style/Markdown rules are inlined in a ~215-line file where a reference file would fit — and there is a minor sequencing slip ('三步:' in 第四步 introduces four numbered items).

4 / 5

Total

18

/

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: it states the complete what and when, lists natural triggers in both Chinese and English including the skill name, documents an optional parameter, and carves out negative scope against sibling skills. The only conceivable criticism is length, but every clause is functional rather than padded.

DimensionReasoningScore

Specificity

The description names four concrete actions covering the full pipeline — '读昨天你与 AI 的全部对话', '照出暴露的思维盲区', '从微信读书挑一本书的一章精准补上', '落成一篇完整分析笔记' — so input, analysis, selection, and output are all explicitly stated with no coverage gaps.

5 / 5

Completeness

Both 'what' (read yesterday's conversations → surface blind spot → pick a targeted weread chapter → write an analysis note) and 'when' (explicit 'Use when user says …' with concrete phrases) are clearly answered, going further with an optional date parameter and negative scope. Matches the top anchor exactly; the when-clause is explicit, not implied.

5 / 5

Trigger Term Quality

Comprehensive natural trigger coverage in both languages — '扫盲区', '盲区', '照盲区', '看看我的思维盲区', '我昨天想漏了什么', 'blind spot', plus the skill name and slash form 'ljg-blind' / '/ljg-blind', and an English intent paraphrase. Nothing a user would naturally say is missing.

5 / 5

Distinctiveness Conflict Risk

Clear niche (cognitive blind-spot scan of conversation history plus a complementary weread chapter) with distinct triggers, and it explicitly disambiguates against four sibling skills via the NOT FOR clause (ljg-think, ljg-rank, ljg-read, ljg-constraint), minimizing wrong-skill triggering.

5 / 5

Total

20

/

20

Passed

Validation

87%

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

Validation — 14 / 16 Passed

Validation for skill structure

CriteriaDescriptionResult

frontmatter_unknown_keys

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

Warning

relative_links

Relative link issues: 1 missing

Warning

Total

14

/

16

Passed

Repository

lijigang/ljg-skills

Path

skills/ljg-blind/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.