Content
32%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 skill is an auto-generated data dump of repetitive regime patterns with no bundle files, no validation workflow, and weak actionability. It functions as a reference table rather than actionable guidance, and its bulk should be externalized.
Suggestions
Move the per-regime pattern entries into a separate data file (e.g. references/regimes.json or REGIMES.md) and keep SKILL.md as a concise overview that links to it, lifting progressive disclosure and conciseness.
Deduplicate and consolidate the repeated regime sections into one canonical entry per regime (Moderate Bull, Trending Up, etc.) instead of 30 overlapping copies.
Replace the vague 4-step usage list with an explicit, validated workflow: identify regime via specific price-action/volatility criteria, select the matching strategy, define risk/position limits, then define a re-evaluation trigger to detect regime changes.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is a 250+ line dump of ~30 near-identical regime entries with heavily duplicated labels (e.g. 'Moderate Bull' repeated 8 times, 'Trending Up' 10 times) and verbose, repetitive blockquotes; this is noticeably padded rather than the lean overview a SKILL.md should be. | 2 / 5 |
Actionability | Each regime offers a semi-concrete directive (e.g. '<25 trades/24h', 'avoid shorting entirely', 'prefer BNB'), which is more than abstract description, but there are no executable steps, commands, or a clear procedure to apply them. | 3 / 5 |
Workflow Clarity | A 4-step 'How to Use' sequence exists but steps are vague ('Adjust your trading approach accordingly') and there are no validation checkpoints for a financially consequential activity; the missing-validation cap and poorly defined steps keep this at the 'rough sequence, many gaps' anchor. | 2 / 5 |
Progressive Disclosure | No bundle files exist and the entire regime-pattern dataset is inlined as one monolithic block that clearly belongs in a separate data file; section headers alone do not compensate for the lack of file separation or navigation. | 2 / 5 |
Total | 9 / 20 Passed |