manus

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

• Third-party content exposure detected (high risk: 0.90). The skill sends prompts to and fetches task outputs/files from the third‑party Manus API (see references/api.md and the runtime client scripts, e.g., scripts/manus_client.mjs cmdResult and collectAssistantOutputs which read assistant text and download file URLs from task output), so untrusted/user-generated content from Manus or connector-backed sources is ingested and can directly influence subsequent decisions or actions.