CtrlK
BlogDocsLog inGet started
Tessl Logo
SkillPublic

ljg-writes

写作引擎。把一个观点写成可理解、可迁移、经得住反例的 1000-1500 字中文文章。USE WHEN 写文章 OR 优化思想内容 OR 展开观点 OR 改写成逻辑递进的中文。NOT FOR 普通摘要、事实查询或结构风洞(用 ljg-structure)。

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

71%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 a highly specific, actionable writing methodology with concrete output constraints, concrete save/verify commands, and a coherent workflow including validation checkpoints. Its weaknesses are moderate redundancy across sections (the same governance rules restated two to three times) and a routing table whose referenced workflow file is absent from the bundle, leaving the skill's core detailed workflow unverifiable.

Suggestions

Ship the referenced "Workflows/WriteEssay.md" alongside SKILL.md (or inline its essential steps) — the routing table currently points to a file that does not exist in the bundle, breaking the skill's stated execution flow.

Deduplicate restated rules: state the internal-checks rule (概念/关系/案例/迁移 stay internal) and the user-length override once each instead of repeating them across 成文契约, Gotchas, and 输出约束.

Add a recovery action for the post-save validation step (e.g., what to fix and retry if the Denote read-back verification fails) to close the workflow's validation loop.

DimensionReasoningScore

Conciseness

The body is dense, prescriptive craft guidance with no padding explaining concepts Claude already knows, but the same rules are restated across sections — the internal-checks rule appears twice ("以上四项用于内部检查,不直接用作文章目录" and "「概念、关系、案例、迁移」等检查项目留在内部") and the user-length override appears in both Gotchas and 输出约束 ("按用户长度交付" / "服从用户长度"). Mostly efficient but could be tightened, matching anchor 3 rather than 4 where over-explanation would be only minor.

3 / 5

Actionability

For an instruction-only skill the guidance is concrete and executable: exact commands ("date +%Y%m%dT%H%M%S"), a copy-paste filename pattern ("~/Documents/notes/{时间戳}==z--{标题关键词}__write.md"), a full YAML file-header template, a hard word-count default (1000-1500 字), and two worked example traces. Specific and directive enough to cover the common cases, matching the top anchor for instruction-only skills.

5 / 5

Workflow Clarity

A clear sequence is present: the analysis-path diagram (案例 -> 原有解释 -> … -> 检查其他场景与边界), a draft-review loop ("初稿完成后,先检查观点与证据,再通读修改中文"), a final read-through check, and an explicit post-save validation ("保存后读回文件,验证 Denote 接受,再报告路径"). Not level 5 because no error-recovery action is specified if the read-back verification fails, and the detailed workflow is delegated to an external file.

4 / 5

Progressive Disclosure

The body has good section structure (成文契约 / 怎样解释 / 姿态与语言 / 最后检查 / Gotchas / 输出约束) and a clearly signaled one-level routing table, but its sole external reference "Workflows/WriteEssay.md" does not exist in the bundle — the routing target is unverifiable/broken, so appropriate content splitting cannot be confirmed. This is worse than level 4's 'minor organization gaps' but better than level 2, since the in-file structure itself is solid and the reference is clearly signaled, not buried.

3 / 5

Total

15

/

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, well-structured description: it states a concrete capability with a measurable output (1000-1500 字中文文章 with named quality properties), provides explicit USE WHEN trigger phrases, and disambiguates against a sibling skill with a NOT FOR clause. The only weaknesses are mildly abstract phrasing in two trigger terms and missing common synonyms like 写作/润色.

DimensionReasoningScore

Specificity

The description names the domain and several concrete actions ("把一个观点写成可理解、可迁移、经得住反例的 1000-1500 字中文文章", "优化思想内容", "展开观点", "改写成逻辑递进的中文") with a concrete output spec. It falls short of level 5 because phrases like "优化思想内容" and "展开观点" are somewhat abstract rather than fully concrete actions, and it sits clearly above level 3's '1-2 concrete actions'.

4 / 5

Completeness

It explicitly answers both what ("把一个观点写成…1000-1500 字中文文章" with named quality properties) and when ("USE WHEN 写文章 OR 优化思想内容 OR 展开观点 OR 改写成逻辑递进的中文") with concrete trigger phrases, plus explicit NOT FOR exclusions — a direct match for the top anchor. Not level 4 because the 'when' clause is already explicit and specific, not merely adequate.

5 / 5

Trigger Term Quality

Trigger phrases "写文章", "改写成逻辑递进的中文" are natural terms a Chinese-speaking user would say when needing this skill. It is not level 5 because common variations such as 写作, 润色, or 写一篇 are missing, and it is well above level 3's 'some relevant keywords'.

4 / 5

Distinctiveness Conflict Risk

A clear niche (Chinese viewpoint-to-essay writing engine) with distinct triggers and an explicit boundary clause "NOT FOR 普通摘要、事实查询或结构风洞(用 ljg-structure)" that delegates the sibling skill, giving minimal conflict risk. Matches the level 5 anchor; only a slight breadth of "写文章" keeps it from being unambiguous, which is already addressed by the exclusions.

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-writes/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.