just-fucking-cancel

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

• Third-party content exposure detected (high risk: 0.90). The skill's workflow (SKILL.md step "5. Cancel Subscriptions") instructs the agent to use browser automation to navigate to cancel URLs from references/common-services.md (external sites like netflix.com, spotify.com, etc.), meaning the agent will fetch and act on public third‑party webpages whose content it must read/interpret and which could therefore influence its actions.