vectorbt

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

• Third-party content exposure detected (high risk: 0.80). The SKILL.md explicitly instructs fetching OHLCV from public third-party sources (e.g., vbt.YFData.download from Yahoo Finance and the birdeye‑api skill for Solana tokens), which are untrusted public data ingested by the workflow and directly influence backtest decisions and parameter selection.