Content
87%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The body is lean, concrete, and well-structured with executable CLI guidance and a clear fan-out workflow; the main gap is the absence of any validation or verification checkpoint for the batch re-render.
Suggestions
Add a validation checkpoint after each format re-render (e.g. verify the subject is centred/uncropped and lighting matches before accepting the asset id, retry on failure) to satisfy the batch-operation feedback-loop expectation.
State how to handle a failed `muapi image edit` call within a node (retry budget, fallback to curl path, or surface error) so the parallel batch degrades gracefully.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Lean Inputs table and numbered Steps assume competence with no concept-explanation padding; every section earns its place, with only a minor redundant restatement of the description under the header. | 3 / 3 |
Actionability | Gives the exact CLI command ("muapi image edit"), model id (nano-banana-2-edit), a complete templated prompt with placeholders, parallel-fan-out instruction, a curl fallback, and explicit asset-id return mapping. | 3 / 3 |
Workflow Clarity | A clear numbered sequence with an explicit skip rule is present, but this batch operation over multiple formats has no validation/verification checkpoint or error-recovery feedback loop, which caps workflow clarity at 2. | 2 / 3 |
Progressive Disclosure | A compact skill with no external references needed, organized into clear labeled sections (Inputs, Steps, Notes, Trigger Keywords), satisfying the simple-skills allowance for a top score. | 3 / 3 |
Total | 11 / 12 Passed |