Content
57%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 concept-rich reference with concrete signal-construction code and a strong output template, but it reads more as an encyclopedic body than an actionable skill: mechanics are over-explained, there is no explicit analysis workflow with validation steps, and all reference material is inlined in a single long file.
Suggestions
Move the ETF category tables, thematic ETF lists, and flow-data source listings into a references/ file and keep SKILL.md as a lean overview with one-level-deep pointers.
Replace the conceptual Creation/Redemption explanation with only the non-obvious interpretation rules, and add an explicit numbered workflow (fetch data → compute breadth/C-D ratio → classify signal → render output) with a validation step for missing or stale flow data.
Make etf_flow_signal self-contained by showing how daily_flows and aum are obtained, and note the 5-day vs 15-day window scale mismatch in the momentum calculation.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The Creation/Redemption mechanics section re-explains AP basket mechanics Claude already knows, and tables state basics like 'XLK | Technology', while the signal-threshold code and data-source table are genuinely additive — mostly efficient but with trimmable explanation across several sections, fitting the 3 anchor better than 4. | 3 / 5 |
Actionability | Provides executable snippets (yahoo_download('SPY', ...)), complete breadth and cyclical/defensive ratio computations, and a copy-paste output template; however, etf_flow_signal depends on undefined daily_flows/aum structures and the flow-data section concedes flows require web sources it never scripts, leaving minor gaps below the 5 anchor. | 4 / 5 |
Workflow Clarity | Sections imply a data → signals → output sequence but no explicit numbered workflow or validation checkpoints (e.g., sanity-checking flows against volume or handling missing flow data) — the sequence is implicit, matching the 3 anchor rather than 4. | 3 / 5 |
Progressive Disclosure | There are no bundle files at all and the ~270-line body inlines large ETF, thematic, and data-source reference tables that belong in separate reference files; headers are well-organized, but content that should be separate is inline, matching the 3 anchor. | 3 / 5 |
Total | 13 / 20 Passed |