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 tightly written, highly actionable skill body: executable commands, a complete CLI table, and a clear four-step workflow with post-run filtering and retry guidance. The only flaws are minor — no error-recovery guidance for failed/timed-out API runs and a script path that doesn't resolve relative to the skill root.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean and operational throughout — commands, a CLI table, filter rules, and presentation guidance with no padding or explanation of concepts Claude already knows; even the Notes (engagement >100%, 1-3 minute runtime) are non-obvious domain facts that earn their tokens. | 5 / 5 |
Actionability | Fully executable, copy-paste-ready bash commands covering the common cases, a complete CLI reference table with defaults and flag semantics, and concrete filtering rules with explicit thresholds ("generally exclude fit < 0.6"). | 5 / 5 |
Workflow Clarity | The four steps (gather criteria → run search → filter → present) are clearly sequenced, with the filtering step acting as a checkpoint and an explicit adjust-and-retry loop ("offer to run another search with different keywords or adjusted criteria"); however there is no guidance for handling run failures or timeouts, leaving a minor validation gap. | 4 / 5 |
Progressive Disclosure | Implementation is appropriately split into the one-level-deep scripts/find_influencers.py, well-signaled via Quick Start, Step 2, and the CLI reference, and the body stays an overview; but the referenced path "skills/tiktok-influencer-finder/scripts/find_influencers.py" is repository-root relative rather than skill-root relative (the bundle file is scripts/find_influencers.py), a minor navigation gap. | 4 / 5 |
Total | 18 / 20 Passed |