two-axis-review

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

• Third-party content exposure detected (high risk: 0.75). The required workflow can fetch outsider-authored issue/PRD text from public or authenticated issue trackers at runtime (via the “issue-tracker workflow” / gh issue view / glab), and then injects that fetched free-form spec text into the Spec sub-agent prompt (step 4. “Spec sub-agent prompt” includes “the path or fetched contents of the spec”).