Content
53%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 impeccably lean and well-sectioned, but it is a shell: every substantive task is delegated to three markdown files that are missing from the bundle, plus an external URL the runtime may not be able to fetch. As shipped, the skill provides no usable guidance for any of its advertised tasks.
Suggestions
Add the three referenced files (get-audio-duration.md, get-video-dimensions.md, get-video-duration.md) to the bundle, or inline a short executable Mediabunny snippet for each task so the body is self-sufficient.
Include at least one copy-paste-ready code example in the body for the most common task, so the skill still works if a reference file is missing.
Replace the external https://mediabunny.dev/llms.txt link with a bundled capabilities reference, since network fetching at skill-load time cannot be assumed.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is a lean 13 lines: one disambiguating sentence ('Mediabunny is a multimedia library for dealing with audio and video in the browser'), a capabilities link, and three one-line pointers, with no padding or explanation of concepts Claude already knows. Every token earns its place, matching the anchor 5 example. | 5 / 5 |
Actionability | The body contains no code, commands, or API usage — only navigation lines like 'See [get-audio-duration.md](get-audio-duration.md) for getting the duration of an audio file in seconds', and those target files do not exist in the bundle. This is minimal concrete guidance with no executable steps (anchor 2), not 3 because nothing runnable is provided and not 1 because the task-level pointers are at least specific. | 2 / 5 |
Workflow Clarity | The three task sections ('Getting audio duration', 'Getting video dimensions', 'Getting video duration') each give one unambiguous pointer, but following the workflow dead-ends because the referenced files are missing and no inline fallback exists. Sequence is present but the workflow cannot actually be completed, matching anchor 3; not 2 because the task-to-file mapping itself is well defined. | 3 / 5 |
Progressive Disclosure | Scored against the actual bundle: get-audio-duration.md, get-video-dimensions.md, and get-video-duration.md are all absent (no references/ or scripts/ directories exist), so the body is an index pointing to nothing, and the capability overview is outsourced to an external llms.txt URL. This is minimal structure with broken references (anchor 2); not 1 because the body itself is well-sectioned with one-level, clearly signaled references rather than nested chains. | 2 / 5 |
Total | 12 / 20 Passed |