conducting-wireless-network-penetration-test

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt instructs capturing and documenting plaintext credentials (including an explicit example "Company2024!") and thus would require the LLM to output secret values verbatim.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). The content explicitly implements and documents offensive wireless attacks — deauthentication (DoS), evil-twin/rogue AP deployment, captive-portal credential harvesting, WPA2/WPA3 handshake/PMKID capture and offline cracking — which are deliberate credential-theft and man-in-the-middle techniques (no obfuscation, backdoor C2, or external exfiltration endpoints are present, but the toolset is fully capable of malicious abuse if used without authorization).

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 1.00). The skill passively and actively ingests untrusted, user-generated wireless frames (Dot11Beacon, Dot11ProbeReq) via sniff()/airodump-ng as shown in scripts/agent.py and SKILL.md and then uses discovered SSIDs/BSSIDs to drive actions like rogue AP/evil-twin creation and targeted attacks, so third-party content can materially influence tool behavior.

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (high risk: 0.90). The skill explicitly instructs running privileged wireless tools (airmon-ng, hostapd-mana, dnsmasq, aireplay-ng) and deploying rogue APs and services that require root, modify network interfaces/configuration, and create/alter service configuration files—actions that change the host machine's state and effectively require privilege escalation.