idea-incubation

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

• Third-party content exposure detected (high risk: 0.70). The required workflow includes a “perform research” step that can fetch external web content and then uses that evidence to generate phase outputs, so outsider-authored free-text from public web pages can be ingested into the agent’s LLM context (e.g., via research findings/evidence blocks), which creates an indirect prompt-injection risk.