acquire

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 shows the agent will use Membrane to run Acquire actions like "list-messages" and "get-case" (see "Popular actions" and the "Running actions" / membrane action run examples), which fetch user-generated/untrusted customer messages from a third-party service and could be read and acted on by the agent.