bundle-social-platform-ops

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

• Third-party content exposure detected (high risk: 0.90). The skill's required workflow (SKILL.md and references/09-misc.md) instructs the agent to call Bundle.social endpoints that import or return user-generated public content—e.g., GET /api/v1/misc/reddit/post-requirements and GET/POST /api/v1/misc/google-business/reviews and Facebook recommendations import—which means the agent will ingest and act on untrusted third-party social media/forum/review content.