Content
25%Weight 40%Scale 1-5Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The body is a well-headed but bloated capability catalog that describes rather than instructs: no runnable code or commands, no sequenced workflows with validation, and eight documented scripts that are missing from the bundle entirely. Its main assets are the concrete platform character limits and JSON input/output schemas, which keep it above the lowest scores.
Suggestions
Either ship the eight documented scripts in scripts/ or remove the Scripts section; right now Claude is told to rely on files that do not exist.
Add executable invocation guidance: how to run each script (command line with arguments) and a worked end-to-end workflow (keyword research -> metadata optimization -> character-limit validation -> tracking) with an explicit validation checkpoint before submitting metadata.
Move the Best Practices, Limitations, and Integration sections into one or two reference files (e.g., references/best-practices.md) and cut the seven duplicate 'Hey Claude' sample prompts to one or two, shrinking SKILL.md to a lean overview with clearly signaled one-level-deep references.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | At ~400 lines the body is noticeably verbose with several padded sections: seven near-identical "Hey Claude..." sample prompts, generic ASO best-practice lists Claude largely already knows (respond to reviews quickly, front-load keywords), and an "Integration with Other Skills" section. Not a 1 because the platform character limits and script API listings are genuine reference content delivered in structured form. | 2 / 5 |
Actionability | There is not a single executable command, code snippet, or invocation example; the JSON input schemas are concrete but nothing shows how to run anything. Critically, the eight documented scripts (keyword_analyzer.py, metadata_optimizer.py, etc.) do not exist in the bundle - no scripts/, references/, or assets/ directory is present - so the function inventories reference files Claude can never actually call. This is high-level description with the execution steps missing, matching anchor 2. | 2 / 5 |
Workflow Clarity | The body is a capability catalog, not a sequenced process: no section gives a step-by-step workflow (e.g., research keywords -> optimize metadata -> validate character limits -> track rankings), and validation checkpoints are absent everywhere despite batch/destructive-adjacent operations like metadata overwrites. A rough implicit sequence (research, optimize, track) exists only at section level, matching anchor 2. | 2 / 5 |
Progressive Disclosure | This is a monolithic ~400-line SKILL.md with zero external reference files, while content that clearly belongs in separate files - API docs for eight scripts, best-practice guides, output format specs - is fully inlined. Worse, the Scripts section points to files that are absent from the bundle, so the only 'references' are dangling. Anchor 2 (content that should be separate is inlined; references buried/broken) is the best fit. | 2 / 5 |
Total | 8 / 20 Passed |