Security Scanning Tools

Warn

Audited by Gen Agent Trust Hub on Mar 10, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill describes workflows involving Nmap, Masscan, Metasploit, Aircrack-ng, and other powerful tools that perform deep network and system interactions.
  • [COMMAND_EXECUTION]: Multiple core operations require elevated privileges using sudo, including service management for Nessus and OpenVAS, enabling wireless monitor mode with airmon-ng, and performing system-wide audits with lynis.
  • [EXTERNAL_DOWNLOADS]: The skill provides instructions for installing external software packages, such as the openvas suite via the system package manager (apt) and cloud security tools like prowler and scoutsuite via the Python package manager (pip).
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8) due to its heavy reliance on processing user-provided targets (IPs, domains, and URLs) alongside its extensive system capabilities.
  • Ingestion points: Target hostnames, IP ranges, and URLs provided as parameters for scanning tools (e.g., nmap, nikto, zap-cli).
  • Boundary markers: No explicit delimiters or instructions are provided to the agent to treat target strings as data and ignore embedded instructions.
  • Capability inventory: Extensive use of sudo, network interaction tools, and software installation capabilities across the entire skill.
  • Sanitization: The skill does not define validation or sanitization steps for the input strings before passing them to the command-line interface.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 10, 2026, 06:51 AM