Content
90%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 strong, dense instruction-only skill: lean tables instead of prose, copy-paste templates, concrete numbers throughout, and an explicit wave-based feedback loop. The only gaps are that validation is framed as template fields rather than enforced checkpoints, and the single-file layout forgoes reference-file splitting at ~175 lines.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is table-driven and dense with non-obvious specifics (budget tiers with dollar figures, "~5–15% will post unprompted", "1.5–3× CTR, 20–60% lower CPI", sourcing-channel pros/cons with costs). It assumes Claude's competence — no explanation of what TikTok or influencer marketing is — and every section earns its tokens. Not 4: there are no padded passages to trim. | 5 / 5 |
Actionability | For an instruction-only skill this is fully executable: a copy-paste-ready creator brief template with concrete guardrails ("Show the app on screen for at least 50% of the video", "6 months paid usage on Meta/TikTok/Google"), a complete output program-plan template, per-tier budgets, and a numbered 5-step paid-amplification workflow. Specific examples cover the common cases end-to-end; not 4 because there are no minor gaps. | 5 / 5 |
Workflow Clarity | A clear sequence is present (Initial Assessment → Strategy Tiers → brief → sourcing → paid workflow → measurement → output template) with an explicit feedback loop ("Scale winners, retire losers, brief next wave"; "Decision criteria for wave 2: what defines a winner"). Not 5: measurement checks and decision criteria are listed as template fields rather than wired into explicit validation gates (e.g., verifying usage rights or disclosure compliance before a wave launches). Not 3: checkpoints are not merely implicit — the wave-1/wave-2 decision loop is spelled out. | 4 / 5 |
Progressive Disclosure | The skill is a single well-sectioned file with clear headers and clearly signaled cross-skill handoffs (`ua-campaign`, `press-and-pr`, `app-clips`, `referral-program`, `localization`); no bundle files exist, so there is no nesting risk and no buried references. Not 5: at ~175 lines, substantial content (the brief template, hook-angle table) could arguably live in reference files to keep SKILL.md a leaner overview. Not 3: organization is genuinely good — everything is well placed and easy to navigate. | 4 / 5 |
Total | 18 / 20 Passed |