linux-pentester-practical-commands

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill includes commands that embed plaintext secrets and keys verbatim (e.g., echo "ssh-rsa YOUR_PUBLIC_KEY" >> authorized_keys, echo 'backdoor:password' | chpasswd, reverse shells with $ATTACKER_IP), which require placing secret values directly into generated commands/files — an exfiltration risk.

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.80). Mixed signals: the GitHub linpeas release and ara.so look legitimate, but the attacker-hosted HTTP file (http://$ATTACKER_IP:8000/file), the generic HTTP file (http://example.com/file), and an unvetted GitHub repo (HIMANSHUSHARMA20/Linux-for-a-Pentester.git) are direct/untrusted download sources commonly used to deliver malware, so the set should be treated as suspicious.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). The content contains explicit, actionable malicious techniques — reverse shells, remote code execution, credential harvesting/exfiltration, persistence mechanisms (cron/systemd/authorized_keys/sudoers), and log-clearing — that enable unauthorized access and backdoors and are therefore highly abusive if misused.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill contains a runtime command that fetches and executes remote code: "curl -L https://github.com/carlospolop/PEASS-ng/releases/latest/download/linpeas.sh | sh", which downloads and pipes a script to sh (executing remote code) and is presented as a tool to run during assessments.

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 creating users, adding SSH keys, modifying /etc/passwd and /etc/sudoers, creating/enabling systemd services, clearing logs, and other actions that require sudo and directly modify the host state, so it clearly pushes the agent to compromise the machine.