prompt-engineer

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

• Third-party content exposure detected (high risk: 0.90). The Prompt Framework's "Learning Journey" / Resource Map explicitly instructs the agent to fetch and load public URLs (e.g., https://protocol-spec.com and GitHub READMEs like https://github.com/.../python-sdk/README.md) as part of Phase 1–2 study, meaning the agent would read untrusted third‑party web content and use it to guide planning and execution.