offensive-krack-fragattacks

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 0.90). This skill contains explicit, actionable instructions for active Wi‑Fi exploitation (deauthentication to force association, rogue AP setup, frame injection and A‑MSDU-based data splicing) that clearly enable interception, credential capture and covert data exfiltration in practice, though it does not contain evidence of backdoors, remote code execution payloads, supply‑chain tampering, or obfuscated malicious code.

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

• Third-party content exposure detected (high risk: 0.80). The SKILL.md workflow explicitly instructs cloning and running public third-party repositories (e.g., "git clone https://github.com/vanhoefm/krackattacks-scripts" and "git clone https://github.com/vanhoefm/fragattacks") and references external sites (krackattacks.com, fragattacks.com), meaning the agent would fetch and act on untrusted public content whose outputs can change testing/reporting decisions.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 1.00). The skill explicitly runs git clone and then executes scripts from https://github.com/vanhoefm/krackattacks-scripts and https://github.com/vanhoefm/fragattacks, so it fetches and executes remote code at runtime and relies on those repos as required dependencies.

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

• Attempt to modify system services in skill instructions detected (high risk: 1.00). The skill explicitly instructs running privileged commands (sudo hostapd, aireplay-ng, krack/fragattacks test scripts) and to perform active network/attack operations that require elevated privileges and modify the host's network state, so it pushes the agent to use sudo and change machine state.