onna

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

• Third-party content exposure detected (high risk: 0.90). The SKILL.md explicitly directs the agent to use the Membrane CLI to query Onna and retrieve "Search Result" and "Source Content" from connected third-party apps (e.g., Slack, Google Drive, Salesforce), so the agent will ingest and act on untrusted user-generated content that could contain embedded instructions.