zk-proofs

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

• Third-party content exposure detected (high risk: 0.80). The SKILL.md's "Source-of-truth checks (required)" explicitly directs the agent to fetch and verify public third‑party pages (e.g., CAP-0059/CAP-0074/CAP-0075 on GitHub and Stellar developer docs at developers.stellar.org), which are untrusted web sources the agent is expected to read and that can materially change its decisions about using ZK primitives.