Content
76%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, information-dense reference: executable code, precise thresholds, careful units conventions, and a concrete output template make it highly actionable. The weakest areas are the lack of an explicitly sequenced analysis workflow with validation checkpoints, and minor redundancy explaining perpetual-futures mechanics Claude already knows.
Suggestions
Add an explicit numbered analysis workflow (fetch funding/OI/basis data → validate units and data freshness → classify regime → evaluate carry and divergence signals → render output) with validation checkpoints, e.g. verifying the funding rate is a per-8h decimal before computing annualized figures.
Trim the funding-rate mechanics primer ('Perpetual futures have no expiry date…') and merge the comment block and units-convention block, which restate the same decimal convention twice.
Consider moving the threshold tables and historical pattern statistics into a reference file to keep the always-loaded SKILL.md leaner.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is dense with skill-specific value (threshold tables, units conventions, OKX endpoint paths, regime thresholds) but includes minor over-explanation of concepts Claude already knows, e.g. 'Perpetual futures have no expiry date… to keep the perpetual price anchored to the spot price.' Not 5 because that mechanics primer and some plain-English restatements (e.g. re-explaining the annualization comment block plus the units-convention block) could be trimmed; not 3 because padding is minor and nearly all content is non-obvious. | 4 / 5 |
Actionability | Guidance is fully executable: runnable functions (funding_regime, annualized_basis, carry_trade_pnl, oi_funding_matrix), concrete numeric examples with worked results ('0.00015 × 3 × 30 × $100,000 = $1,350'), exact OKX API endpoint paths, and a complete output-format template. Not 4 because the common cases (regime detection, basis calc, carry P&L, data access) are all covered with copy-paste-ready code. | 5 / 5 |
Workflow Clarity | The carry-trade execution sequence (steps 1–5) is clearly listed, but there is no explicitly sequenced end-to-end analysis workflow — fetch data → classify regime → compute basis → assess carry/divergence → produce the output format is only implied by section ordering, and validation checkpoints (e.g. sanity-check fetched rates against expected decimal units, confirm both carry legs are equal-sized before entry) are missing. Not 4 because checkpoints are absent rather than a minor gap; not 2 because the execution steps that do exist are well-defined and the output format anchors the end state. | 3 / 5 |
Progressive Disclosure | No bundle files exist (references/, scripts/, assets/ are absent), so the skill is deliberately single-file; it is well-sectioned with clear headers (Core Concepts 1–7, Data Access, Output Format, Notes) and the one cross-skill pointer 'Use load_skill("okx-market")' is clearly signaled and one level deep. Not 5 because at ~250 lines some reference material (e.g. the signal tables and historical pattern statistics) could be split out of the always-loaded overview; not 3 because navigation is easy and nothing is buried. | 4 / 5 |
Total | 16 / 20 Passed |