screenshot-website

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt instructs the agent to extract and then embed the returned session_id into subsequent CLI commands (e.g., /sessions/{session_id}/...), which requires outputting a secret-like token verbatim and thus creates an exfiltration risk.

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

• Third-party content exposure detected (high risk: 0.80). The workflow explicitly navigates to arbitrary public URLs via the Notte browser (see "Step 2: Navigate to the URL" using orth api run notte /sessions/{session_id}/page/execute with a user-supplied "url"), thereby ingesting untrusted third‑party web content as part of its runtime behavior which could influence subsequent actions.