Content
100%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.
An excellent instruction-only skill body: dense with Tavily-specific parameters, empirically-derived caveats, and concrete examples, sequenced as a clear search-verify-extract workflow with explicit validation checkpoints and error-recovery loops. It stays lean, assumes Claude's competence, and appropriately keeps everything in one compact file.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Every sentence carries non-obvious operational knowledge — concrete parameter values, empirically observed source quirks (e.g. 'a country-specific Indeed hostname did not remove unrelated roles'), and decision rules — with no padding or explanations of things Claude already knows. | 5 / 5 |
Actionability | Fully actionable for an instruction-only skill: copy-paste-ready query examples ('Data Engineer jobs Berlin Germany'), literal include_domains lists, explicit settings (search_depth: "advanced", max_results: 10), and a complete extraction prompt template covering the common cases. | 5 / 5 |
Workflow Clarity | Clear four-stage sequence (build searches → screen/verify → extract → research/shortlist) with validation at its core: a 'Check and record' checklist, explicit geography/recency/status verification, and feedback loops for extraction failure (retry canonical page, mark unknown) and time_range widening. | 5 / 5 |
Progressive Disclosure | No bundle files exist and none are needed: the ~53-line body is self-contained, well-organized into four numbered sections plus a run-shape summary, matching the simple-skill exception for short skills with clear section structure. | 5 / 5 |
Total | 20 / 20 Passed |