linux-privilege-escalation

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.90). These links include direct downloads of executable scripts from raw GitHub URLs (exploit/tool PoCs) plus explicit attacker-controlled HTTP endpoints (http://ATTACKER/ and /cron) — direct executable downloads and attacker-hosted URLs are high-risk for malware or tampering.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). The files are an explicit offensive Linux privilege-escalation playbook with step-by-step instructions for gaining root, persistence, credential theft, backdoors, and exfiltration — clearly intended for malicious abuse.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill contains explicit runtime commands that download and execute remote scripts (e.g., curl|sh or wget + chmod && ./) such as https://github.com/peass-ng/PEASS-ng/releases/latest/download/linpeas.sh, https://raw.githubusercontent.com/ly4k/PwnKit/main/PwnKit, https://raw.githubusercontent.com/mzet-/linux-exploit-suggester/master/linux-exploit-suggester.sh, and https://raw.githubusercontent.com/jondonas/linux-exploit-suggester-2/master/linux-exploit-suggester-2.pl which fetch and run external code.

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (high risk: 1.00). The skill explicitly instructs the agent to perform privileged, state-changing actions (adding users, modifying /etc/passwd and /etc/shadow, creating SUID binaries, using LD_PRELOAD with sudo, writing SSH keys via docker, editing cron/systemd jobs, etc.), which directly compromise the host machine.