zendesk-guide

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

• Third-party content exposure detected (high risk: 0.70). This skill explicitly connects to Zendesk Guide via the Membrane CLI and runs actions that fetch and return article/comment data from a third-party knowledge base (see "Connecting to Zendesk Guide" and "Running actions" in SKILL.md), i.e., user-generated content the agent would read and that could materially influence subsequent actions.