market-structure-analyzer

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly fetches and ingests live data from open public third‑party sources (OKX HTTP endpoints and CLI, OnchainOS CLI, CoinMetrics, CoinGecko, Alternative.me, DefiLlama, and optional Dune queries) as shown in SKILL.md and Data Sources, and it uses that untrusted external content to compute composite signals and drive analysis, so those third‑party responses could materially influence the agent's decisions and enable indirect prompt injection.