yield-analysis

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

• Third-party content exposure detected (high risk: 1.00). The skill's scripts and documentation explicitly fetch and parse live public DeFi data (e.g., scripts/yield_comparison.py calls the DeFiLlama API at https://yields.llama.fi/pools and SKILL.md lists DeFiLlama and protocol APIs), so untrusted third‑party content is ingested and directly influences ranking and decision logic.