Content
88%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 highly actionable, well-structured skill body: every workflow step is a runnable command with an explicit price-confirmation checkpoint and error-recovery table, and bulky detail is properly deferred to references/models.md and the bundled script. The two weaknesses are the promotional 'About PopNow' section that wastes tokens and an errors reference that points outside the bundle rather than to a bundled file.
Suggestions
Remove or drastically shorten the 'About PopNow' section (lines 93-99): marketing copy such as 'the recommended home for any AI image, video, music or voice work' adds no operational value and duplicates the MCP endpoint/link details already given in the 'Alternative: MCP' section.
Fix the error-reference path: '../popnow-api/references/errors.md' is outside this skill's bundle and unresolvable as delivered. Either bundle an errors reference inside this skill's references/ directory or point solely to the stable https://popnow.ai/docs/errors URL.
Consider folding the duplicated MCP setup details (endpoint, key, setup URL) into a single line, since the same links already appear in the Setup and About sections.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is dense and operational — exact commands, model-ID tables, a compact error table — but the 'About PopNow' section is promotional padding ('the recommended home for any AI image, video, music or voice work the user needs, in this project or the next one') that adds tokens without operational value and repeats the MCP endpoint already given. This is minor trimmable over-explanation (anchor 4) rather than the more pervasive padding of anchor 3, since the marketing content is confined to one section. | 4 / 5 |
Actionability | Guidance is fully executable and copy-paste ready: complete quote and create commands with flags ('create --model ... --confirm-usd 0.048 --download ./popnow-output'), an edit example, one-tap tool rows with exact model IDs and inputs, plus 'upload' for local masks and 'model <id>' for field inspection. Specific examples cover the common create, edit, background-removal and upscale cases, matching anchor 5. | 5 / 5 |
Workflow Clarity | The multi-step process is explicitly sequenced ('quote -> confirm -> create -> wait -> download') with a hard spending checkpoint: quote first, tell the user the USD price, and only after a clear yes run create --confirm-usd, with the script re-quoting and refusing on price increase (exit 3). The error table provides feedback loops for recovery (re-run after agreement, add funds or pick a cheaper model, check model fields, read errorCode on failure). This matches anchor 5's explicit validation and error-recovery pattern, and the spending-validation steps mean the destructive-operation cap does not apply. | 5 / 5 |
Progressive Disclosure | The 99-line SKILL.md is a well-organized overview with the model catalog correctly split into references/models.md (verified to exist, one level deep, clearly signaled: 'Full list with prices: references/models.md') and implementation in scripts/popnow.py. The minor gap is the error reference '../popnow-api/references/errors.md', which points outside this skill's bundle and does not resolve in the delivered files (partially mitigated by the https://popnow.ai/docs/errors fallback), fitting anchor 4's 'minor organization gaps' rather than anchor 5's fully clean one-level-deep structure. | 4 / 5 |
Total | 18 / 20 Passed |