paper-writing

Workflow 3: Full paper writing pipeline that goes from a narrative report to a polished, submission-ready PDF. Use when user says "写论文全流程", "write paper pipeline", "从报告到PDF", "paper writing", or wants the complete paper generation workflow.

Quality

85%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

81%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 highly actionable, well-sequenced pipeline with strong validation gates, feedback loops, and concrete commands throughout. Its main weakness is verbosity from repeated boilerplate (style-ref forwarding, assurance derivation) that could be consolidated or factored into referenced files.

Suggestions

Consolidate the repeated 'If — style-ref was passed and the helper succeeded above, append ...' instruction into a single stated-once rule referenced by each phase, instead of restating it ~5 times.

Factor the assurance-level derivation (currently duplicated in Phase 0, 5.9, and 6.0) into one canonical block and have the other phases reference it.

Move the large Phase 1.5 contract-negotiation and Phase 5.9 forensics-gate prose into separate reference files linked from SKILL.md to reduce the monolithic inline bulk.

Dimension	Reasoning	Score
Conciseness	The body is operationally dense and does not explain basic concepts Claude already knows, but it is noticeably over-long: the style-ref forwarding instruction is restated verbatim ~5 times and the assurance-level derivation is repeated across Phases 0, 5.9, and 6.0, so it could be meaningfully tightened.	3 / 5
Actionability	It provides copy-paste-ready slash-command invocations ('/paper-plan "$ARGUMENTS"', '/paper-figure "PAPER_PLAN.md"'), concrete bash blocks (style-helper resolution, numeric-claim rg detector, verifier resolution), and a full Codex MCP prompt covering the common cases.	5 / 5
Workflow Clarity	The pipeline is explicitly sequenced (Phase 0 → 1 → 1.5 → 2 → 2b → 3 → 4 → 4.5 → 4.7 → 5 → 5.5 → 5.6 → 5.8 → 5.9 → 6) with inter-phase checkpoints, a literal submission pre-flight checklist, and feedback loops (improvement loop review→fix→recompile, contract negotiation up to 3 rounds, fix-and-revalidate).	5 / 5
Progressive Disclosure	It has clear phase headers and clearly signaled one-level-deep references (links to ../shared-references/*.md and ../integrity-forensics/SKILL.md), but large inlined policy blocks (contract negotiation, forensics gate) that could live in separate files keep it just short of a 5; no bundle files exist under references/scripts/assets to back the in-skill references.	4 / 5
	Total	17 / 20 Passed

Description

90%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, third-person description that clearly states what the skill does and when to use it, with rich natural trigger terms in both English and Chinese. Its only gap is that it describes the pipeline as a single end-to-end transform rather than naming the constituent sub-actions.

Dimension	Reasoning	Score
Specificity	It names the domain ('paper writing pipeline') and one concrete end-to-end action ('from a narrative report to a polished, submission-ready PDF'), but does not enumerate the several sub-actions (plan, figures, write, compile, improve) the body actually performs, so it is not comprehensive.	3 / 5
Completeness	It explicitly answers both 'what' ('Full paper writing pipeline that goes from a narrative report to a polished, submission-ready PDF') and 'when' ('Use when user says ... or wants the complete paper generation workflow') with concrete trigger phrases.	5 / 5
Trigger Term Quality	It lists multiple natural trigger phrases a user would actually say in two languages — '写论文全流程', 'write paper pipeline', '从报告到PDF', 'paper writing', 'complete paper generation workflow' — covering synonyms and bilingual variants comprehensively.	5 / 5
Distinctiveness Conflict Risk	It carves a clear niche (narrative-report → submission-ready PDF) with distinct bilingual trigger phrases, giving it minimal overlap risk with other skills.	5 / 5
	Total	18 / 20 Passed

Validation

75%

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 — 12 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (913 lines); consider splitting into references/ and linking	Warning
allowed_tools_field	'allowed-tools' contains unusual tool name(s)	Warning
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning
relative_links	Relative link issues: 7 suspicious	Warning

	Total	12 / 16 Passed

Repository: wanshuiyin/Auto-claude-code-research-in-sleep
Path: skills/paper-writing/SKILL.md
Commit: f4f20f9

Reviewed: 1 day 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.