Content
56%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.
The body is well-organized with genuinely valuable non-obvious specifics (trigger timing numbers, conversion benchmarks, frequency caps, compliance and Google interstitial notes) and a clear assessment-to-output workflow. However, roughly a third of it restates UX/copy knowledge Claude already possesses, and the monolithic 450-line SKILL.md inlines reference material that should be split into bundle files.
Suggestions
Cut the "Core Principles" section and trim "Design Best Practices" and "Copy Formulas" to only non-obvious specifics (trigger timing numbers, sizing ranges, mobile exit-intent limitation, first-person CTA finding) — Claude already knows generic popup/UX conventions.
Trim the ~90-line "Experiment Ideas" section down to a handful of high-value tests per popup type rather than an exhaustive bullet enumeration of variations Claude can generate on demand.
Split reference material (Popup Types details, Experiment Ideas, Common Popup Strategies) into separate bundle files (e.g., references/popup-types.md, references/experiments.md) with one-level-deep pointers, keeping SKILL.md focused on the assessment workflow and output format.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Of ~450 lines, roughly a third restates knowledge Claude already has — "Core Principles" ("Too early = annoying interruption", "Don't trap or trick"), generic design conventions (close button placement, touch targets), and a ~90-line Experiment Ideas section enumerating trivial variations ("Test with/without images", "Button color testing"). This is noticeably verbose with several padded sections. | 2 / 5 |
Actionability | Concrete, specific guidance throughout — "Show after 30-60 seconds", "25-50% scroll depth", "Desktop: 400-600px wide", "7-30 days before showing again", benchmarks ("Email popup: 2-5% conversion typical"), copy examples, and a defined Output Format. Falls short of 5 only because there is no complete worked example of a full popup specification. | 4 / 5 |
Workflow Clarity | The process has an explicit entry point (read product-marketing-context.md, then "Before providing recommendations, understand: …" and Task-Specific Questions) and an explicit exit point (Output Format, Test Hypotheses), with the middle organized as topical reference. This is advisory, non-destructive work, so no validation checkpoints are required. | 4 / 5 |
Progressive Disclosure | Section headers are clear, but all ~450 lines live inline in SKILL.md with no bundle files; Popup Types deep-dives, Design Practices, Common Popup Strategies, and Experiment Ideas are reference material that clearly belongs in separate files, matching the anchor for structure present but content that should be separate is inline. | 3 / 5 |
Total | 13 / 20 Passed |