scout

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

• Third-party content exposure detected (high risk: 0.85). Outsider-authored free text from public sources—e.g., README/docs, issue bodies, and other peer repository content fetched via GitHub search and releases APIs at runtime in Step 2 (“Every finding carries a locator… Each agent writes full findings… and returns a summary; synthesize from the files”)—is then summarized and placed into the agent/LLM context, enabling indirect prompt injection.