catalan-adobe/gemini-icon-set
Generate production-quality colorful icon sets using Google Imagen 4. Use this skill when the user needs custom app icons, emoji-style icons, cartoon icons, or any colorful icon set. Takes a project description, suggests icons, generates 1024px PNGs via Imagen 4, removes backgrounds, downscales to all target sizes, and delivers a review gallery for iterative refinement. Trigger on: "icon set", "generate icons", "app icons", "custom icons", "colorful icons", "cartoon icons", "emoji icons", "make icons", "icon generation", "imagen icons", or any request for a set of colorful visual assets.
74
92%
Does it follow best practices?
Impact
—
No eval scenarios have been run
Passed
No known issues
Quality
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 an excellent skill description that hits all the marks. It provides specific concrete actions in a clear pipeline, includes comprehensive natural trigger terms, explicitly states both what the skill does and when to use it, and occupies a distinct niche. The description is thorough without being unnecessarily verbose.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: suggests icons, generates 1024px PNGs via Imagen 4, removes backgrounds, downscales to target sizes, and delivers a review gallery for iterative refinement. Very detailed pipeline description. | 3 / 3 |
Completeness | Clearly answers both 'what' (generate production-quality icon sets, remove backgrounds, downscale, deliver review gallery) and 'when' (explicit 'Use this skill when...' clause plus a detailed 'Trigger on:' list). | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural trigger terms including 'icon set', 'generate icons', 'app icons', 'custom icons', 'colorful icons', 'cartoon icons', 'emoji icons', 'make icons', and 'icon generation'. These are terms users would naturally use. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive with a clear niche: colorful icon set generation specifically using Google Imagen 4. The combination of icon-specific triggers and the specific tool (Imagen 4) makes it very unlikely to conflict with other image generation or design skills. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
77%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a well-structured, highly actionable skill with an excellent iterative workflow and clear validation checkpoints via the gallery review loop. Its main weaknesses are moderate verbosity (the script-locating fallback, duplicated pipeline overview, and some explanatory notes Claude doesn't need) and inline content that could benefit from being split into reference files. The concrete commands, JSON schemas, and precise file manipulation instructions make this very executable.
Suggestions
Remove or significantly shorten the script-locating fallback block — the primary `CLAUDE_SKILL_DIR` approach is sufficient, and Claude can figure out fallback strategies on its own.
Remove the 'Pipeline Overview' ASCII diagram since it duplicates the detailed Execution section that follows immediately after.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Generally efficient but includes some unnecessary detail. The script-locating fallback block is verbose, the pipeline overview duplicates the step-by-step execution section, and some notes (like explaining what Lanczos resampling does) are things Claude already knows. However, most content earns its place. | 2 / 3 |
Actionability | Highly actionable with concrete bash commands, exact JSON manifest format, specific file paths, and copy-paste ready code blocks for every step. The iteration workflow (delete specific files, re-run pipeline) is precise and executable. | 3 / 3 |
Workflow Clarity | Excellent multi-step workflow with clear sequencing (Steps 0-7), explicit validation via gallery review, and a well-defined feedback loop (Steps 5-6 repeat until satisfied). The retry/drop/add mechanics are clearly specified with exact file operations for each case. | 3 / 3 |
Progressive Disclosure | References a styles file at `references/STYLES.md` which is good progressive disclosure, but no bundle files were provided to verify it exists. The SKILL.md itself is fairly long (~200 lines) and some content like the standalone installation section and detailed script-locating fallback could potentially be in separate reference files. The overall structure is well-organized with clear headers though. | 2 / 3 |
Total | 10 / 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.
Reviewed
Table of Contents