shortcut

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

• Third-party content exposure detected (high risk: 0.80). The skill's SKILL.md explicitly instructs extracting IDs from URLs under https://app.shortcut.com/ and running the short CLI or short api to fetch and display story/epic details (user-generated content), which the agent will read and could let embedded instructions influence subsequent actions.