Content
85%Reviews 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-structured, highly actionable skill with clear workflows, validation checkpoints, and excellent progressive disclosure to verified reference files. Its main weakness is conciseness: the prompt-only caveat is restated several times and the incident anecdote could be trimmed.
Suggestions
State the "image_generate is prompt-only and does not accept reference images" caveat once in the Reference Images section and reference it elsewhere rather than repeating the full explanation in Step 7, 7.1, 7.2, and Pitfalls.
Condense the "Incident Apr 2026" anecdote into a one-line caution; the actionable rule ("always pass an absolute path to curl -o") already stands on its own.
Consider moving the lengthy clarify-timeout handling detail into references/workflow.md and keeping a brief pointer in the main body.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is largely operationally dense and avoids explaining concepts Claude already knows, but it repeats the "image_generate is prompt-only / does not accept images" point across Reference Images, Step 7, 7.1, 7.2, and Pitfalls, and the "Incident Apr 2026" anecdote is padded narrative, fitting score 2 ("Mostly efficient but could be tightened") rather than score 3's "every token earns its place." | 2 / 3 |
Actionability | It provides copy-paste-ready commands ("curl -fsSL \"<url>\" -o /abs/path/...\"), an aspect-ratio mapping table, exact prompt-file naming conventions, and the image_generate schema, matching the score-3 anchor "Fully executable code/commands; specific examples; copy-paste ready." | 3 / 3 |
Workflow Clarity | It includes a progress checklist, a flow diagram, a step-summary table, conditional review gates, and explicit validation checkpoints ("Verify the file exists and is non-empty ... before proceeding"), plus feedback loops for clarify timeouts, matching the score-3 anchor for clear sequencing with validation and error-recovery loops. | 3 / 3 |
Progressive Disclosure | SKILL.md is a concise overview pointing one level deep to verified-real reference files (art-styles/, tones/, presets/, layouts/, workflow.md, auto-selection.md, partial-workflows.md) with clear load signals ("Full definitions at ...", "load the file when a preset is picked"), matching the score-3 anchor for well-signaled one-level-deep references and easy navigation. | 3 / 3 |
Total | 11 / 12 Passed |