bland-ai

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

• Third-party content exposure detected (high risk: 0.80). This skill invokes Membrane actions to fetch user-created/untrusted Bland AI data (e.g., "Get Call Details", "List Calls", "List Web Agents", "List Pathways" in SKILL.md) which return transcripts, pathways, and tool configurations that the agent is expected to read and could materially influence subsequent actions.