skills/reason-machines/security-skills/wireless-security-wifi-penetration-testing/Gen Agent Trust Hub
wireless-security-wifi-penetration-testing
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFECOMMAND_EXECUTIONPRIVILEGE_ESCALATIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides numerous shell commands for network interface manipulation, packet injection, and cryptographic attacks using the aircrack-ng suite and other utilities.
- [PRIVILEGE_ESCALATION]: Extensive use of
sudois required for many operations, as these tools require direct interaction with wireless hardware and system network configurations. - [EXTERNAL_DOWNLOADS]: The skill instructs the user to clone the
eaphammertool from a third-party GitHub repository (s0lst1c3/eaphammer) and execute its setup script. - [INDIRECT_PROMPT_INJECTION]: The skill defines a workflow where the agent processes untrusted wireless network metadata (such as SSIDs, BSSIDs, and EAP identities) gathered by tools.
- Ingestion points: Output from tools like
airodump-ng,wash, andkismetcontaining data from the local RF environment. - Boundary markers: No explicit markers or warnings are used to prevent the agent from interpreting instructions potentially hidden in wireless metadata (e.g., a malicious SSID).
- Capability inventory: The skill possesses high-privilege capabilities including shell execution, hardware control, and package installation.
- Sanitization: No sanitization of external wireless data is mentioned before it is processed by the agent.
Audit Metadata