serpdog

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

• Third-party content exposure detected (high risk: 0.80). The SKILL.md instructs the agent to call Serpdog search actions via Membrane (e.g., "Search / Search Results" and using "membrane action run" to get the action output), which returns public/search-result content from third‑party websites that the agent is expected to read and act on, exposing it to untrusted open-web content.