Content
87%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The content is concise and highly actionable with concrete CLI guidance and a clear two-phase workflow, held back only by the absence of explicit verification checkpoints after image generation.
Suggestions
Add an explicit checkpoint after Phase A and Phase B instructing the agent to confirm the generated/edited image is valid (e.g. poll with 'muapi predict wait' and verify a non-empty image URL) before proceeding.
Note the expected output artifact (image URL/path) at each phase so the agent knows what a successful result looks like.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean and instruction-focused with no preamble explaining interior design concepts; prompts, model IDs, and commands are given directly and every line earns its place. | 3 / 3 |
Actionability | It provides concrete executable commands (muapi image generate / image edit) with model IDs, full parameterized prompts, aspect ratios, and a curl-plus-polling fallback — copy-paste ready. | 3 / 3 |
Workflow Clarity | The two phases are clearly sequenced with the dependency made explicit, but there are no explicit validation/polling checkpoints after each generation step to confirm the image succeeded before proceeding. | 2 / 3 |
Progressive Disclosure | A short (~50 line) single-file skill with well-organized sections (Inputs, phased Steps, Trigger Keywords, Notes) and no bundle files, which meets the simple-skill threshold for a top score. | 3 / 3 |
Total | 11 / 12 Passed |