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.

Quality

56%

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

Fix and improve this skill with Tessl

tessl review fix ./SKILLs/canvas-design/SKILL.md

The canonical home for this skill is canvas-design in anthropics/skills

Quality

Content

35%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 lays out a coherent creative workflow with organized sections and instructive examples, but it is verbose and lacks concrete executable guidance (no libraries, commands, or rendering code) despite producing code-generated artifacts. Validation checkpoints and progressive disclosure via bundle files are underdeveloped.

Suggestions

Trim repeated craftsmanship/minimal-text injunctions into a single concise principles list to reduce padding and respect the token budget.

Add concrete, executable guidance for producing .png/.pdf output — e.g., a named rendering library, a code skeleton, or specific commands — instead of only 'Download and use whatever fonts are needed.'

Move the five full philosophy examples and detailed canvas-generation guidance into a reference file under references/, and add an explicit verify/refine checkpoint (e.g., render, inspect for overlaps/margins, fix, re-render) to the FINAL STEP loop.

Dimension	Reasoning	Score
Conciseness	The body is noticeably verbose and padded — phrases like 'The philosophy MUST stress multiple times', repeated injunctions to emphasize craftsmanship, and restatements of the same minimal-text principle appear many times without adding new depth.	2 / 5
Actionability	Despite the skill producing code-based output (.pdf/.png), the body offers almost no concrete, executable guidance — it instructs to 'Search the ./canvas-fonts directory' and 'Download and use whatever fonts' but gives no commands, libraries, rendering code, or concrete tooling, leaving execution largely abstract.	2 / 5
Workflow Clarity	A rough two-step sequence (philosophy creation, then canvas creation) plus a final refinement pass is present, but it lacks concrete validation checkpoints for a generative/destructive output step; the FINAL STEP refinement loop is described abstractly rather than as a verify-then-fix cycle.	3 / 5
Progressive Disclosure	Sections are organized with headers, but the entire workflow is inlined into one ~100-line SKILL.md with no bundle files in references/, scripts/, or assets/ — content like the five full philosophy examples and detailed canvas guidance could be split into separate reference files for clearer navigation.	3 / 5
	Total	10 / 20 Passed

Description

78%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 clearly covers both what the skill does and when to use it, with natural trigger terms and explicit file extensions. Its main weakness is specificity — the actions named ('create art/design') are generic rather than a comprehensive list of concrete operations.

Suggestions

Replace generic verbs like 'create visual art' with a more comprehensive enumeration of concrete operations (e.g., 'generate posters, abstract compositions, typographic layouts, and pattern-based visuals').

Consider framing in third person ('Creates...' / 'Use when...') instead of second person ('You should use this skill') to align with the rubric's voice guidance.

Dimension	Reasoning	Score
Specificity	It names concrete actions ('Create beautiful visual art', 'Create original visual designs', output '.png and .pdf documents') but the actions are generic ('create art/design') rather than enumerated specific operations, leaving coverage non-comprehensive.	3 / 5
Completeness	It explicitly answers both 'what' (create visual art in .png/.pdf using design philosophy) and 'when' with concrete trigger phrases ('You should use this skill when the user asks to create a poster, piece of art, design, or other static piece').	5 / 5
Trigger Term Quality	It includes natural user phrases ('create a poster, piece of art, design, or other static piece') plus file extensions (.png, .pdf); a few common synonyms like 'illustration' or 'graphic' are missing but coverage is solid.	4 / 5
Distinctiveness Conflict Risk	The focus on static visual art (.png/.pdf) via design philosophy is a fairly distinct niche with clear triggers, though 'design' broadly could overlap with general document-formatting or brand skills.	4 / 5
	Total	16 / 20 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
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: netease-youdao/LobsterAI
Path: SKILLs/canvas-design/SKILL.md
Commit: 100d3f7

Reviewed: 2 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.