Content
42%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.
The body delivers genuinely useful crypto-specific content — thresholds, parameter suggestions, a decision tree, and an output template — but buries it in textbook derivatives explanations and non-executable snippets, all in one monolithic file. Trimming known concepts and splitting reference material into bundle files would substantially raise the score.
Suggestions
Delete or drastically compress sections explaining concepts Claude already knows (Basic Greeks table, contango/backwardation definitions, funding-payment mechanics), keeping only the crypto-specific deltas like 'BTC IV is often 50-120%' and 'crypto trades 7x24'.
Make the term-structure snippet executable: import pandas, show how prices are fetched, or explicitly label it as an illustrative formula and provide a runnable script in scripts/.
Move detailed reference material (strategy parameter tables, exchange comparison, historical regularities, output template) into references/*.md files and link them one level deep from SKILL.md.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | A large fraction of the body restates textbook knowledge Claude already has: the funding mechanism ('longs pay shorts'), 'Contango: far month > near month > spot', and a full Basic Greeks table defining Delta/Gamma/Theta/Vega. The crypto-specific specifics (settlement times, IV bands, thresholds) are the real value, but several sections are unnecessary explanation, matching 'noticeably verbose' rather than 'mostly efficient'. | 2 / 5 |
Actionability | Concrete elements exist — funding-rate signal bands, strategy parameter tables ('Strike: 10-15% below spot', 'IV > 80% consider selling'), a decision tree, and an output template. But the sole code sample is illustrative rather than executable (uses pd.Timestamp without importing pandas, no data retrieval), and data acquisition is hand-waved ('available through the OKX data source'), so guidance is incomplete rather than copy-paste ready. | 3 / 5 |
Workflow Clarity | The Analysis Framework section sequences the work (monitoring metrics → market-environment judgment via decision tree → strategy selection → output format), which is a coherent flow. However, the steps from raw data to the output table are implicit, and there are no validation checkpoints (e.g., verifying fetched prices or sanity-checking computed basis), matching 'sequence present but checkpoints missing'. | 3 / 5 |
Progressive Disclosure | The body has a clear section hierarchy (perpetuals, term structure, options, framework, output) but is a single ~280-line file with no bundle at all, so detailed reference material (strategy parameter tables, exchange characteristics, the output template) that belongs in separate files is inlined. Structure exists, but the split is not made. | 3 / 5 |
Total | 11 / 20 Passed |