Content
70%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
A well-structured, mostly actionable skill body with clear sequencing and good progressive disclosure. The two weaknesses are a non-existent 'cross-chain' subcommand documented as executable, and minor command duplication between Instructions and Examples.
Suggestions
Either implement the 'cross-chain' subcommand in arb_finder.py or remove the cross-chain instruction and example to keep all documented commands executable.
Replace the duplicated command blocks in Examples with references to the Instructions (e.g., reuse step 4) and instead show distinct sample output or edge cases to improve token efficiency.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The body is lean and assumes Claude's knowledge (no 'what is arbitrage' padding), but the Examples section re-prints three commands already shown in Instructions (scan ETH USDC, triangular binance, cross-chain USDC), which could be tightened. | 2 / 3 |
Actionability | Most commands are concrete and executable with verified flags, but the documented 'cross-chain' subcommand (Instructions item 5 and the cross-chain Example) does not exist in arb_finder.py, which only implements scan/triangular/monitor/calc, so one of seven invocations is non-functional. | 2 / 3 |
Workflow Clarity | Seven numbered, labeled steps form an unambiguous sequence, and the Error Handling table provides explicit recovery guidance (stale prices -> retry, rate limited -> reduce polling or add API key); this is a read-only scanning skill so the destructive/batch feedback-loop cap does not apply. | 3 / 3 |
Progressive Disclosure | The body is a concise overview pointing to one verified, clearly signaled reference (references/implementation.md) one level deep, with content appropriately split into Overview/Prerequisites/Instructions/Output/Errors/Examples/Resources. | 3 / 3 |
Total | 10 / 12 Passed |