analyze-crypto

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

• Third-party content exposure detected (high risk: 1.00). The skill's Phase 1 subagents explicitly instruct the agent to perform WebSearch and WebFetch on open web sources—e.g., price sites (CoinGecko/CoinMarketCap), news originals, "crypto twitter" community posts, and the project website (see Agent 1, Agent 2, and Agent 5 prompts in SKILL.md)—and then requires the main thread to read, interpret, cross-correlate, and act on those third‑party results, so untrusted public content could materially influence behavior and enable indirect prompt injection.