audn-quick-scan

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

• Third-party content exposure detected (high risk: 1.00). This skill explicitly registers and scans a user-supplied agent API URL (Step 1 uses "api_url": "$ARGUMENTS"), executes campaigns against that endpoint (Step 4), and fetches/parses the target agent's responses via "response_path" and result endpoints (Steps 4–5), meaning it ingests untrusted third‑party agent output and uses it to drive vulnerability findings and decisions.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill fetches attack scenarios at runtime from the audn.ai API (base URL https://audn.ai via GET /api/scenarios) which directly supply prompt payloads used to test the target agent, so the external content controls prompts and is a required runtime dependency.