linux-pentester-command-reference
Fail
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: CRITICALCOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONDATA_EXFILTRATIONPRIVILEGE_ESCALATIONPERSISTENCECREDENTIALS_UNSAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: Provides one-liners for reverse shells using Bash, Python, Netcat, and FIFOs.
- [REMOTE_CODE_EXECUTION]: Instructs the agent to download scripts from external GitHub repositories and execute them locally using chmod +x.
- [DATA_EXFILTRATION]: Details methods for exfiltrating files and directory archives to an external IP using Netcat.
- [PRIVILEGE_ESCALATION]: Includes techniques for SUID binary exploitation, Sudo vulnerability abuse, kernel exploits, and environment PATH hijacking.
- [PERSISTENCE]: Documents techniques to maintain system access via shell configuration files, cron jobs, and SSH authorized keys.
- [CREDENTIALS_UNSAFE]: Instructs the agent to search for and read sensitive files such as password hashes in /etc/shadow and SSH private keys.
- [EXTERNAL_DOWNLOADS]: References external tools and code repositories from third-party sources like GitHub and HackTricks.
- [DYNAMIC_EXECUTION]: Employs python -c and bash -c to execute dynamically constructed shell and socket commands.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted system output and network data which could be manipulated to influence agent behavior. 1. Ingestion points: System command outputs and target network information. 2. Boundary markers: Absent; instructions do not distinguish between command output and new instructions. 3. Capability inventory: Full shell command execution, file system modification, and network capabilities. 4. Sanitization: No input validation or verification of external script integrity.
Recommendations
- CRITICAL: 2 file(s) identified as malware by FileRep - DO NOT USE
- AI detected serious security threats
- Contains 2 malicious URL(s) - DO NOT USE
Audit Metadata