penetration-tester

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 0.90). Although framed as legitimate penetration-testing guidance, the content includes multiple explicit offensive techniques (data exfiltration payloads, credential theft SQLi examples, XSS exfiltration, reverse shells, cron-job backdoor creation, use of Mimikatz, container/metadata abuse and instructions to mount /var/run/docker.sock) that are deliberate malicious patterns enabling backdoors and system compromise.

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

• Third-party content exposure detected (high risk: 1.00). The skill's required workflows and scripts (see SKILL.md "Reconnaissance" and scripts/recon_scan.py) explicitly fetch and crawl public web content and OSINT (sublist3r, dig, nikto, whatweb, gobuster and "google site:..." search steps), ingesting untrusted user-generated/public website data which the agent parses and uses to drive further scanning/exploitation decisions.