shortcut

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

• Third-party content exposure detected (high risk: 0.80). The skill parses and fetches story/epic data from app.shortcut.com (via URL parsing and commands like short story <id> and short epic view <id>), which ingests user-generated/untrusted Shortcut content (titles, descriptions, comments) that the agent will read and present.