Content
50%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The skill body is rich and well-structured with concrete code, tables, and an output template, but it is verbose, relies on pseudocode-style helpers, and lacks an explicit validated workflow and progressive file split.
Suggestions
Tighten or move conceptual explanation (liquidation mechanics, cascade anatomy, exchange-level differences) into a reference file to reduce token load, since Claude already knows these basics.
Make the Python examples self-contained by defining or stubbing helper functions (e.g. estimate_long_liq_at_price) and passing price_range as a parameter.
Add an explicit numbered analysis workflow with validation checkpoints (e.g. verify data source coverage, cross-check OI changes) before producing the output.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The ~280-line body is mostly efficient with useful tables and code, but spends tokens explaining concepts (liquidation mechanics, cascade anatomy, exchange differences) that Claude already knows and could be tightened. | 2 / 3 |
Actionability | It provides concrete Python functions and formulas, but they depend on undefined helpers (estimate_long_liq_at_price) and an unpassed price_range variable, making them incomplete rather than copy-paste ready. | 2 / 3 |
Workflow Clarity | Content is organized by topic with a useful before/during/after cascade guide, but there is no explicitly sequenced analysis workflow with validation checkpoints for the overall process. | 2 / 3 |
Progressive Disclosure | It is a single monolithic file with well-organized sections but no external references, and substantial reference-style content (metrics tables, exchange comparisons) is inlined rather than split into separate files. | 2 / 3 |
Total | 8 / 12 Passed |