Content
87%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
A tight, actionable reference with executable recipes and good organization; its main gap is the absence of explicit validation/feedback checkpoints, though that is partly excused by the non-destructive nature of animation work.
Suggestions
Add a short validate/iterate step for ScrollTrigger setups (e.g. call ScrollTrigger.refresh() after load and verify trigger firing) to give the workflow an explicit feedback loop.
Consider splitting the 'Key concepts & APIs' reference into a separate REFERENCE.md if the file grows, keeping SKILL.md as an overview pointing one level deep.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean: compact API signatures, pitfalls as terse problem->fix lines, and three focused recipes, with no padding explaining what GSAP is or basic programming concepts Claude already knows; every token earns its place. | 3 / 3 |
Actionability | It provides concrete, executable, copy-paste-ready code (hero stagger, sequenced timeline, pinned scroll-scrub) plus exact API signatures and stagger/ease syntax, with specific pitfall fixes rather than vague direction. | 3 / 3 |
Workflow Clarity | Recipes are concrete single-purpose sequences and the 'What to ask the user' section acts as a requirements checkpoint, but there are no explicit validate->fix->retry feedback loops or sequenced multi-step guidance with checkpoints, which the rubric expects for a 3; not capped lower because animation work is non-destructive so the missing-validation cap does not apply. | 2 / 3 |
Progressive Disclosure | No references/scripts/assets bundle exists, so all content is in a single well-organized file with clear sections (When to use, Key concepts, Common pitfalls, Quick recipes); there are no nested or deep references to penalize. | 3 / 3 |
Total | 11 / 12 Passed |