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.
58
70%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Advisory
Suggest reviewing before use
Optimize this skill with Tessl
npx tessl skill review --optimize ./skills/paper-illustration/SKILL.mdQuality
Discovery
100%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is a strong skill description that clearly communicates its purpose, lists concrete capabilities, and provides explicit trigger guidance in both Chinese and English. It uses third-person voice consistently and carves out a distinct niche (AI-generated academic illustrations via Gemini). The description is concise yet comprehensive, covering what, when, and how it differs from generic diagram or image generation skills.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: 'Generate publication-quality AI illustrations', 'Creates architecture diagrams', 'method illustrations', and mentions 'Claude-supervised iterative refinement loop'. These are concrete, specific capabilities. | 3 / 3 |
Completeness | Clearly answers both 'what' (generate AI illustrations for academic papers using Gemini, creates architecture diagrams with iterative refinement) and 'when' (explicit 'Use when' clause with specific trigger phrases and the general condition 'needs visual figures for papers'). | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural trigger terms in both Chinese and English: '生成图表', '画架构图', 'AI绘图', 'paper illustration', 'generate diagram', 'visual figures for papers'. These are terms users would naturally say. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive niche: academic paper illustrations using Gemini image generation with Claude-supervised refinement. The combination of academic context, specific tool (Gemini), and bilingual trigger terms makes it very unlikely to conflict with other skills. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
39%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The skill demonstrates excellent workflow design with a clear multi-stage iterative process and strong validation checkpoints, but is severely undermined by extreme verbosity and repetition. The same style guidelines, color palettes, and visual standards are repeated 3-4 times throughout the document, inflating it to an unreasonable token cost. The bash scripts, while concrete, have quoting issues that would prevent direct execution, and the monolithic structure makes it difficult to navigate.
Suggestions
Extract the CVPR/NeurIPS style guide, review checklist template, and prompt template into separate referenced files (e.g., STYLE_GUIDE.md, REVIEW_TEMPLATE.md, PROMPT_TEMPLATE.md) to reduce the main file by ~60%
Eliminate the 3-4x repetition of color palettes, visual appeal guidelines, and arrow standards — define them once and reference them
Fix the bash heredoc quoting issues (triple-quoted Python strings inside bash heredocs with variable interpolation) to make scripts actually executable
Remove bilingual repetition where Chinese and English say the same thing — pick one language per guideline or use the non-English only for terms that lack precise English equivalents
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Extremely verbose at ~500+ lines with massive repetition. The CVPR style guide is repeated nearly verbatim in the prompt template, the review checklist, and the style section. Visual appeal guidelines appear at least 3 times. The ASCII workflow diagram, while informative, is oversized. Much content (what a PDF is, how curl works, basic JSON parsing) assumes Claude doesn't know these things. The bilingual Chinese/English repetition doubles content without adding value. | 1 / 3 |
Actionability | Contains executable bash scripts with actual API calls and concrete code, which is good. However, the scripts use placeholder patterns like '$INITIAL_PROMPT' and '[Claude fills in the detailed prompt here]' that aren't truly copy-paste ready. The heredoc quoting with triple-quotes inside bash is broken (mixing Python triple-quotes with bash heredocs). The prompt template has many bracketed placeholders that require significant interpretation. | 2 / 3 |
Workflow Clarity | The multi-step workflow is exceptionally clear with explicit sequencing (Steps 0-8), a visual flowchart, explicit validation checkpoints (score ≥ 9 gate), feedback loops (review → fix → re-validate), and clear decision points (accept vs refine). The iteration loop with MAX_ITERATIONS=5 and specific scoring criteria for pass/fail provides robust error recovery. | 3 / 3 |
Progressive Disclosure | Monolithic wall of text with no references to external files for detailed content. The entire CVPR style guide, prompt templates, review checklists, and bash scripts are all inline. The style guide alone could be a separate reference file. The review template, prompt template, and improvement prompt template each could be separate files. No bundle files are provided to offload this content, resulting in an overwhelming single document. | 1 / 3 |
Total | 7 / 12 Passed |
Validation
72%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 8 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
skill_md_line_count | SKILL.md is long (734 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 | 8 / 11 Passed | |
- Commit
a425a71
- 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.