canvas-design
Create beautiful visual art in .png and .pdf documents using design philosophy. You should use this skill when the user asks to create a poster, piece of art, design, or other static piece. Create original visual designs, never copying existing artists' work to avoid copyright violations.
94
51%
Does it follow best practices?
Impact
98%
1.75xAverage score across 10 eval scenarios
Failed to scan
The risk profile of this skill
Optimize this skill with Tessl
npx tessl skill review --optimize ./skills/anthropic-canvas-design/SKILL.mdQuality
Discovery
67%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 adequately communicates both what the skill does and when to use it, earning good marks on completeness. However, it suffers from somewhat vague language ('design philosophy', 'beautiful visual art') and could benefit from more specific capability listing and broader trigger term coverage. The second-person voice ('You should use this skill') should be penalized per the rubric guidelines, though the rest uses appropriate framing.
Suggestions
Replace vague phrases like 'using design philosophy' and 'beautiful visual art' with specific capabilities (e.g., 'Creates posters, illustrations, infographics, banners, and other static visual designs').
Expand trigger terms to include common user phrases like 'illustration', 'graphic', 'infographic', 'flyer', 'banner', 'visual artwork', 'image creation'.
Rewrite in third person voice ('Creates visual art...') instead of second person ('You should use this skill') to align with style guidelines.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Names the domain (visual art/design) and some actions (create posters, art, designs), but 'using design philosophy' is vague and the specific capabilities like file format handling aren't well elaborated beyond mentioning .png and .pdf. | 2 / 3 |
Completeness | Clearly answers both 'what' (create visual art in .png and .pdf documents) and 'when' ('when the user asks to create a poster, piece of art, design, or other static piece'), with explicit trigger guidance. | 3 / 3 |
Trigger Term Quality | Includes some natural trigger terms like 'poster', 'piece of art', 'design', and file formats '.png' and '.pdf', but misses common variations users might say such as 'illustration', 'graphic', 'infographic', 'flyer', 'banner', 'visual', or 'artwork'. | 2 / 3 |
Distinctiveness Conflict Risk | The mention of 'static piece' and visual art helps distinguish it from general document creation, but 'design' is broad and could overlap with UI design, web design, or presentation skills. The .pdf output could also conflict with PDF-focused skills. | 2 / 3 |
Total | 9 / 12 Passed |
Implementation
35%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The skill provides a creative and ambitious framework for generating visual art through a design philosophy approach, with useful examples of philosophy styles. However, it is severely undermined by extreme verbosity and repetition (especially around craftsmanship emphasis), lacks concrete executable code for the actual canvas generation step, and provides mostly abstract/aspirational direction rather than actionable technical guidance for producing PDFs and PNGs.
Suggestions
Cut repetitive emphasis on craftsmanship, museum quality, and minimal text to single mentions each — the current version repeats these concepts 5+ times without adding new information.
Add concrete, executable code examples showing how to generate a PDF or PNG canvas (e.g., using reportlab, Pillow, or cairo), including specific commands for font loading from `./canvas-fonts`.
Move the philosophy examples into a separate reference file (e.g., PHILOSOPHY_EXAMPLES.md) and link to it, keeping SKILL.md as a concise overview.
Add explicit validation checkpoints: e.g., verify canvas dimensions, check text doesn't overflow boundaries, validate PDF output opens correctly before the refinement pass.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Extremely verbose and repetitive. The instruction to emphasize 'expert craftsmanship' is itself repeated at least 5-6 times across different sections. Concepts like 'minimal text,' 'visual expression,' and 'museum quality' are restated redundantly throughout. The skill could be condensed to roughly 1/3 its length without losing any actionable information. | 1 / 3 |
Actionability | Provides some concrete guidance (philosophy examples, structural approach, font directory reference), but lacks executable code or specific tool commands for generating PDFs/PNGs. The canvas creation section is almost entirely abstract direction ('push aesthetics to the frontier') rather than concrete steps with specific libraries, code, or commands. | 2 / 3 |
Workflow Clarity | The two-step workflow (philosophy creation → canvas creation) is clearly sequenced, and there's a refinement pass at the end. However, there are no validation checkpoints, no error recovery steps, and the 'deducing the subtle reference' step is vaguely defined. The final refinement step lacks concrete criteria for what 'pristine' means in practice. | 2 / 3 |
Progressive Disclosure | Content is structured with clear headers and sections, which helps navigation. However, it's monolithic — all content is inline in one large file with no references to external files (except the font directory). The philosophy examples could be in a separate reference file, and the multi-page option could be split out. | 2 / 3 |
Total | 7 / 12 Passed |
Validation
100%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 11 / 11 Passed
Validation for skill structure
No warnings or errors.
- Repository
- boisenoise/skills-collections
- Commit
431bfad
- 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.