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

78%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./skills/ljg-writes/SKILL.md

Quality

Content

65%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 dense, well-organized instruction skill with genuinely executable output constraints (commands, filename pattern, org template) and explicit validation steps. Its main defects are that the routing table delegates the core workflow to Workflows/WriteEssay.md, which is absent from the bundle, and that this broken reference undermines both workflow clarity and progressive disclosure.

Suggestions

Ship Workflows/WriteEssay.md alongside SKILL.md (or inline its content) — the routing table's '执行时完整读取工作流' currently points to a file that does not exist in the bundle, leaving the core workflow unresolvable.

Convert 最后检查 into an explicit ordered checklist with a hard validation gate before saving, so the editorial validation loop is as concrete as the existing save-time checks (org-lint, Denote read-back).

Consolidate overlapping style rules in 姿态与语言 and Gotchas (several bullets cover the same 'keep meaning, avoid formula' concern) to trim token cost without losing any directive.

Dimension	Reasoning	Score
Conciseness	The body contains no padding or explanations of concepts Claude already knows — every section is a specific directive (e.g., "换近义词仍然别扭，就按整句意思重新组织"). The 9-bullet 姿态与语言 section and the Gotchas list have minor overlap that could be trimmed, placing it just below the lean anchor 5.	4 / 5
Actionability	The output constraints are copy-paste ready: exact shell commands ("date +%Y%m%dT%H%M%S"), an exact filename pattern, and a literal org header template, plus a concrete analysis-path diagram and two worked examples. The core writing guidance is instruction-only but specific, so it mostly meets anchor 4 with minor gaps — the main execution sequence is delegated to an external workflow file.	4 / 5
Workflow Clarity	A sequence with real validation checkpoints exists ("保存后读回文件，验证 Denote 接受与 org-lint，再报告路径"; "初稿完成后，先检查观点与证据，再通读修改"), matching anchor 3's 'steps listed, checkpoints partly present'. It cannot score 4 because the routing directive "执行时完整读取工作流" points to Workflows/WriteEssay.md, which is not present in the bundle — the primary workflow is unresolvable as shipped.	3 / 5
Progressive Disclosure	Sections within the single file are well organized and the one external reference (Workflows/WriteEssay.md) is clearly signaled in a routing table, one level deep. However, no bundle directories exist and that referenced file is missing, so the declared structure does not match the actual bundle — a broken reference places it at anchor 3 rather than 4.	3 / 5
	Total	14 / 20 Passed

Description

83%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 is strong: it states a concrete, measurable deliverable, provides an explicit USE WHEN clause with natural trigger phrases, and adds a NOT FOR clause that disambiguates against a sibling skill. Its only weakness is that the listed actions are variations of a single core task and a few natural synonyms (e.g., 写一篇, 改稿) are missing.

Dimension	Reasoning	Score
Specificity	"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" names a concrete deliverable spec (type, length, qualities) and "改写成逻辑递进的中文" adds a second concrete action. It falls between anchor 3 (1-2 concrete actions) and anchor 5 (multiple distinct actions with comprehensive coverage), since the actions are variations of one core writing task rather than a broad action list.	4 / 5
Completeness	It explicitly answers both questions: the "what" is a viewpoint turned into a 1000-1500 character Chinese essay with named qualities (可理解、可迁移、经得住反例), and the "when" is an explicit USE WHEN clause with concrete trigger phrases, plus a NOT FOR clause. This matches the anchor-5 example structure (action list + 'Use when' with concrete triggers).	5 / 5
Trigger Term Quality	"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" provides four natural phrases a user would actually say. Coverage is good but a few natural synonyms are missing (e.g., 写一篇/帮我写, 改稿), so it does not reach the comprehensive-synonym coverage of anchor 5.	4 / 5
Distinctiveness Conflict Risk	The "NOT FOR 普通摘要、事实查询或结构风洞（用 ljg-structure）" clause explicitly disambiguates against a sibling skill and the niche (viewpoint-driven Chinese essays) is fairly distinct. However, the trigger "写文章" is a broad phrase that could still overlap with generic writing/summarization skills, so minor overlap risk remains — anchor 4 rather than 5.	4 / 5
	Total	17 / 20 Passed

Validation

93%

Warnings & errors only

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

Criteria	Description	Result
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
Commit: 7449831

Reviewed: 2 days ago

Table of Contents

Discovery Implementation Validation

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.