triage

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

• Third-party content exposure detected (high risk: 0.85). The required workflow explicitly reads the full body/comments/labels/diff of an issue or PR (outsider-authored text from the issue tracker) in Step 1 “Gather context,” which is then used to drive the agent/LLM’s reasoning—i.e., indirect prompt injection via issue/PR content.