Content
51%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 is highly actionable with executable code and commands, but it is verbose and repetitive, inlines what should be a separate API reference, and lacks an explicit workflow with validation checkpoints for a destructive batch-trading skill.
Suggestions
Split the bulk TypeScript API reference and chat-command catalog into a separate references file (e.g. REFERENCE.md) and keep SKILL.md as a concise overview with one-level-deep links.
Add an explicit end-to-end workflow with validation checkpoints (verify wallet/funds -> follow with small size -> confirm trade_copied events -> monitor stop-loss -> review daily P&L) since this skill places real trades.
Cut redundancy: sizing modes, stop-loss, take-profit, and daily limits are each shown in chat commands, createCopyTradingService defaults, and dedicated sections — keep one canonical example per concept.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body inlines a large, repetitive API reference — Sizing Modes and Risk Controls are demonstrated three times (chat commands, API defaults, dedicated sections) — and re-explains routine concepts like 'Fixed: Always trade same dollar amount', which is padding Claude does not need. | 2 / 5 |
Actionability | It provides copy-paste-ready TypeScript snippets and concrete chat commands covering the common cases (follow/unfollow, sizing modes, stop-loss/take-profit, top-trader discovery, analysis), with specific option names and values throughout. | 5 / 5 |
Workflow Clarity | Sections are thematically ordered and a start/monitor/stop sequence is implied, but there is no explicit ordered workflow with validation checkpoints; for a destructive/batch trading skill the guidelines cap workflow_clarity at 3 when validation/verification steps are absent. | 3 / 5 |
Progressive Disclosure | No bundle files exist, yet ~180 lines of detailed API reference and chat-command tables that clearly belong in a separate reference file are inlined directly in SKILL.md, with no one-level-deep references or navigation pointers to split the content. | 2 / 5 |
Total | 12 / 20 Passed |