paper-illustration

Generate publication-quality AI illustrations for academic papers using Gemini image generation. Creates architecture diagrams, method illustrations with Claude-supervised iterative refinement loop. Use when user says "生成图表", "画架构图", "AI绘图", "paper illustration", "generate diagram", or needs visual figures for papers.

Quality

76%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

High

Do not use without reviewing

Fix and improve this skill with Tessl

tessl review fix ./skills/paper-illustration/SKILL.md

Quality

Content

58%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 skill defines a strong, well-validated iterative workflow with executable API code, but it is heavily padded with repeated bilingual guidance and large inlined code/prompt blocks that should be split into bundle files. Conciseness and progressive disclosure are the weakest dimensions.

Suggestions

Extract the repeated bash/python API-call blocks (Steps 2–4) and the ~80-line prompt template into scripts/ and references/ files, keeping SKILL.md as an overview that points to them one level deep.

De-duplicate the "thick arrows / no rainbow / print-friendly" style guidance, which currently appears in at least five sections, into a single CVPR/NeurIPS style reference.

Remove or compress the ASCII workflow diagram and the bilingual restatements to cut tokens while preserving the Step 0–8 sequence and the score-gate validation logic.

Dimension	Reasoning	Score
Conciseness	The body is noticeably verbose: a large ASCII workflow diagram, an ~80-line verbatim prompt template, three near-identical bash/python code blocks repeating curl-and-JSON boilerplate, and bilingual "thick arrows / no rainbow" guidance restated across Core Design, the CVPR guide, the prompt template, Step 3, the Step 5 checklist, and Key Rules.	2 / 5
Actionability	Provides concrete, mostly executable bash/python code with real model names, real Gemini API URLs, JSON payload construction, and base64 image extraction; minor gaps such as the "[Claude fills in the detailed prompt here]" placeholder in Step 2 keep it just short of fully copy-paste ready.	4 / 5
Workflow Clarity	A clearly sequenced Step 0–8 multi-stage process with an explicit score >= 9 gate, a STRICT verification checklist as the validation checkpoint, and a targeted feedback loop (Step 6 decision + Step 7 improvement prompt) for error recovery.	5 / 5
Progressive Disclosure	No bundle files exist (references/scripts/assets absent), so all content is inlined in one monolithic file; the large prompt template and the three code blocks that clearly belong in separate script files are inlined rather than split out, and external path references are not clearly signaled as part of this skill's bundle.	2 / 5
	Total	13 / 20 Passed

Description

95%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 well-crafted: it states concrete capabilities in third person and provides an explicit, bilingual "Use when" clause with rich natural trigger phrases. Its only limitation is a slightly modest action list rather than an exhaustive enumeration.

Dimension	Reasoning	Score
Specificity	Names the domain plus several concrete actions ("Generate publication-quality AI illustrations", "Creates architecture diagrams, method illustrations", "Claude-supervised iterative refinement loop"), but the action list is not as comprehensive as a 5-anchor listing five-plus discrete capabilities.	4 / 5
Completeness	Explicitly answers both "what" (generate publication-quality illustrations, architecture diagrams, method illustrations) and "when" ("Use when user says ... or needs visual figures for papers") with concrete trigger phrases, matching the 5-anchor.	5 / 5
Trigger Term Quality	Comprehensive natural trigger phrases in both Chinese and English ("生成图表", "画架构图", "AI绘图", "paper illustration", "generate diagram") plus "needs visual figures for papers", covering synonyms and phrasings a user would actually say.	5 / 5
Distinctiveness Conflict Risk	A clear niche (academic paper illustrations via Gemini image generation) with distinct bilingual trigger terms and minimal overlap risk with other skills.	5 / 5
	Total	19 / 20 Passed

Validation

81%

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 — 13 / 16 Passed

Validation for skill structure

Criteria	Description	Result
skill_md_line_count	SKILL.md is long (737 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

	Total	13 / 16 Passed

Repository: wanshuiyin/Auto-claude-code-research-in-sleep
Path: skills/paper-illustration/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.