arbitrage
Warn
Audited by Snyk on Feb 20, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 1.00). The skill explicitly ingests live market data and user-generated market questions/prices from public prediction platforms (Polymarket, Kalshi, Manifold, Metaculus, PredictIt, Betfair, Smarkets, etc.) as shown in SKILL.md and the index.ts workflows (checkArbitrage, compareMarkets, autoMatchMarkets), and it interprets that content to detect/act on arbitrage and auto-match markets—meeting the conditions for untrusted third-party content influencing agent actions.
Audit Metadata