paper-write

Draft LaTeX paper section by section from an outline. Use when user says "写论文", "write paper", "draft LaTeX", "开始写", or wants to generate LaTeX content from a paper plan.

Quality

88%

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

Quality

Content

77%Weight 40%Scale 1-3

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A thorough, actionable paper-drafting skill with a strong sequenced workflow and executable examples, weakened by redundancy between its Key Rules/Writing Quality Reference sections and inline content plus references to non-existent bundle paths.

Suggestions

Remove the redundant 'Key Rules' and 'Writing Quality Reference' sections, or collapse them to pointers, since they restate details already covered in Steps 4, 5, and 7.

Provide the referenced `templates/` venue files and `shared-references/` protocol files, or remove the links, so navigation is not broken.

Turn the Python bib-cleaning block into runnable code (or move it to a `scripts/filter_bib.py` reference) instead of comment-only pseudocode.

Dimension	Reasoning	Score
Conciseness	The body is dense and mostly assumes Claude's competence, but the "Key Rules" section restates points already made (one-file-per-section, clean bib, backup, flexible section count, page-limit rule) and "Writing Quality Reference" repeats the Step 5 de-AI patterns, so it could be tightened.	2 / 3
Actionability	It provides copy-paste-ready LaTeX template blocks, executable DBLP/CrossRef curl commands, and a concrete MCP review_start call with polling instructions; the only gap is the Python bib-cleaning block, which is comments/pseudocode rather than runnable code.	3 / 3
Workflow Clarity	Steps 0-8 are clearly sequenced with explicit validation checkpoints: backup-before-overwrite in Step 0, the DBLP/CrossRef/[VERIFY] fallback chain in Step 4, bounded review polling in Step 6, and a full final checklist in Step 8.	3 / 3
Progressive Disclosure	References to `templates/` and `../../shared-references/*.md` are well-signaled and one level deep, but those paths do not exist in the bundle, and large blocks (section-specific guidelines, bib verification, de-AI patterns) are inlined in a ~340-line file rather than split into reference files.	2 / 3
	Total	10 / 12 Passed

Description

100%Weight 40%Scale 1-3

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 well-crafted description that pairs a concrete capability statement with explicit bilingual trigger guidance, closely matching the rubric's good examples. It uses third-person/imperative voice with no over-claims.

Dimension	Reasoning	Score
Specificity	"Draft LaTeX paper section by section from an outline" names a concrete action with a specific method (section-by-section, from an outline), paralleling the good example "Generate descriptive commit messages by analyzing git diffs"; it is not vague.	3 / 3
Completeness	It states both what ("Draft LaTeX paper section by section from an outline") and when ("Use when user says...") with explicit triggers, satisfying the what-AND-when bar.	3 / 3
Trigger Term Quality	"写论文", "write paper", "draft LaTeX", "开始写" give good bilingual coverage of the natural phrases a user would actually say, beyond bare technical terms.	3 / 3
Distinctiveness Conflict Risk	The LaTeX section-by-section drafting niche with these specific triggers is clearly distinguishable and unlikely to fire for unrelated skills.	3 / 3
	Total	12 / 12 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
relative_links	Relative link issues: 3 suspicious	Warning

	Total	15 / 16 Passed

Repository: wanshuiyin/Auto-claude-code-research-in-sleep
Path: skills/skills-codex-claude-review/paper-write/SKILL.md
Commit: fe5963c

Reviewed: 17 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.