The Agent Skills Directory

[COMMAND_EXECUTION]: The scripts/agent.py file uses subprocess.run to execute various wireless security tools including airmon-ng, airodump-ng, aircrack-ng, wash, and reaver to automate pentesting tasks.
[PRIVILEGE_ESCALATION]: The instructions in SKILL.md include commands such as sudo bettercap and airmon-ng check kill, which require administrative privileges to interact with hardware interfaces and manage system processes.
[INDIRECT_PROMPT_INJECTION]: The skill processes data from external wireless networks by parsing scan results in scripts/agent.py and scripts/process.py.
Ingestion points: Network metadata is parsed from CSV files generated by airodump-ng in both Python scripts.
Boundary markers: The scripts do not use explicit delimiters or instructions to ignore potential injection within network names (SSIDs).
Capability inventory: The agent has the capability to execute shell commands and write files to the local system.
Sanitization: The parsing logic relies on simple comma separation and does not strictly validate or sanitize SSID content before it is included in generated reports or assessment findings.

performing-wireless-network-penetration-test