poweruser-feature-audit

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

• Third-party content exposure detected (medium risk: 0.30). The skill reads a user-supplied PR URL and then explicitly has a subagent read “the PR and return a fact sheet,” including the “PR body, linked issues, and comments” (untrusted input), which is outsider-authored free text injected via the GitHub/PR content the workflow consumes at runtime.