Content
71%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.
A well-structured, self-contained recipe with strong executable detail (specific model, parameters, and a complete image-generation prompt). The main weaknesses are the verbatim duplicate description and the implicit async polling step, both of which are easy fixes.
Suggestions
Remove the verbatim duplicate of the description below the H1 title — the frontmatter already carries it.
Move the async wait/poll step ('muapi predict wait <request_id>') into the Phase A step sequence so the generation → wait → present flow is explicit.
Show the assembled muapi command line with {{person_image}} substituted so the common case is copy-paste ready.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body repeats the frontmatter description verbatim and includes padded asides ('a generic studio headshot will give weaker results', the IG/TikTok re-cropping rationale), so while the prompt block itself is justified, the content could be tightened. | 3 / 5 |
Actionability | Concrete guidance throughout — 'muapi image edit (model=gpt-image-2-image-to-image)' with exact parameters (2160x3840, png, quality auto, moderation low), a complete copy-paste prompt, and a curl fallback with 'muapi predict wait <request_id>'. Falls short of 5 only because a single fully-assembled runnable command line with substituted placeholders is not shown. | 4 / 5 |
Workflow Clarity | The simple sequence is clear — ask for the photo if missing, generate, present with variation suggestions — but the async wait/poll step sits in the Notes section rather than the step sequence, and no checkpoint confirms the result before presenting it to the user. | 4 / 5 |
Progressive Disclosure | The skill is self-contained (~55 lines, no bundle files exist), with well-organized sections (Inputs, Steps, Trigger Keywords, Notes) and nothing that belongs in external files; per the rubric's simple-skill exception this earns full marks. | 5 / 5 |
Total | 16 / 20 Passed |