hunt

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). This skill contains explicit, actionable instructions for data exfiltration (sending cookies/DOM to attacker-controlled servers, OOB callbacks), credential theft (SSRF to 169.254.169.254 for cloud metadata, OAuth redirect capture), WAF-evasion guidance, obfuscated payloads (eval(atob(...))) and templates for remote code execution — all clear patterns enabling unauthorized compromise and malicious abuse.

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

• Third-party content exposure detected (high risk: 1.00). The skill explicitly fetches and ingests content from arbitrary public targets (e.g., "curl -sI https://$TARGET" in Phase 2 and numerous curl-based tests/SSRF/OOB probes in Phase 3, plus instructions to use external OOB services like interactsh/webhook.site), treats those untrusted responses as inputs the agent must read/interpret, and uses them to drive decisions and subsequent tool actions (WAF preamble, vuln-class selection, test matrices), so it clearly exposes the agent to untrusted third‑party content that could carry indirect prompt injection.