pre-build-feature-audit

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

• Third-party content exposure detected (high risk: 0.90). The required workflow explicitly instructs reading untrusted repository collaboration surfaces (issues/PRs/wikis/comments) in full before classifying overlap, which at runtime can cause outsider-authored free text from those surfaces to be ingested into the agent’s LLM context (even though it says to treat it as untrusted evidence).