campaignhq

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

• Third-party content exposure detected (high risk: 0.80). The SKILL.md shows the agent uses Membrane to fetch and proxy data from CampaignHQ (e.g., via "membrane action run" and "membrane request" and listed actions like Get All Contacts/Get Campaign), which pulls user-generated/untrusted campaign content (contacts, scripts, call history, etc.) that the agent is expected to read and could materially influence subsequent actions.