prediction-arbitrage-scout

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.80). This skill fetches public, user-generated market data from Polymarket and Kalshi (see scripts/fetch_polymarket.py and scripts/fetch_kalshi.py) and then reads and uses the event_name fields in scripts/detect_arbitrage.py and scripts/generate_report.py, so it clearly ingests untrusted third‑party content that could carry indirect prompt injections.