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.
The body is a strong operational skill document: fully executable commands, explicit prerequisites, a clearly sequenced workflow with validation and error-recovery loops, and a done-criteria checklist. Weaknesses are minor — slight repetition of the wrapper framing and some server-side spec detail that could be trimmed.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean and assumes Claude's competence — inputs, defaults, and commands are delivered via compact tables with no explanation of known concepts. Minor trimming is possible: the wrapper/delegation framing is restated in the intro, the comparison table, and Step 3, and the virality-framework list ("hook moments, emotional peaks, opinion bombs…") is server-side spec detail Claude cannot act on, which places it between anchors 4 and 5 but noticeably above the midpoint. | 4 / 5 |
Actionability | Fully executable throughout: a complete `bash library/social/youtube-shorts/scripts/run-youtube-shorts.sh` invocation with flags, prereq commands (`muapi auth configure`, `MUAPI_API_KEY`), quick patterns for defaults/TikTok/square-feed/batch-xargs/async-jq, and a concrete JSON output schema with a real example. These cover the common cases copy-paste ready. | 5 / 5 |
Workflow Clarity | Clear sequence (collect inputs → verify prerequisites → run pipeline) with explicit validation checkpoints (prereq check, "If --output-json was set, the file exists and parses", working clip_urls), feedback loops for error recovery in Failure Modes ("bump --poll-timeout and retry", "re-fetch the same request_id… rather than re-clipping"), and a Done Criteria checklist — matching the anchor's validate → fix → retry structure. | 5 / 5 |
Progressive Disclosure | Scored against the actual bundle structure: no references/, scripts/, or assets/ files exist, so the skill is a single ~160-line file. Section organization is good (tables, quick patterns, schema, failure modes, done criteria) and everything inline is appropriate at this length, but there is no reference split at all — the anchor-5 example of well-signaled one-level-deep references cannot apply, so this sits at good-structure-with-no-split. | 4 / 5 |
Total | 18 / 20 Passed |