building-red-team-c2-infrastructure-with-havoc

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt embeds explicit plaintext operator passwords inside the Havoc profile and instructs creating configuration files that include them, which would require the LLM to output secret values verbatim and poses high exfiltration risk.

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.90). Although hosted on reputable platforms (GitHub, a blog, Microsoft), the majority of these links are to the Havoc C2 framework and red-team deployment guides which are dual-use tooling explicitly intended to build/serve C2 payloads and thus represent a high-risk source for distributing malware.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). This skill explicitly documents how to build and operate a command‑and‑control (C2) infrastructure (payload generation, HTTPS redirectors, Teamserver/listeners), task remote Demon agents to run shells and in‑memory tools (mimikatz, Rubeus, SharpHound), perform token/credential theft and lateral movement, and includes OPSEC/evasion advice (redirectors, domain fronting, sleep/jitter, process injection, indirect syscalls, encrypted payloads) — capabilities that enable remote code execution, backdoors, credential exfiltration and clear abuse despite the “authorized” framing.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly runs a runtime installation step that clones and builds external code from GitHub (git clone https://github.com/HavocFramework/Havoc.git), which fetches and executes remote code that the skill depends on—this satisfies the criteria for a high-risk external dependency.

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

• Attempt to modify system services in skill instructions detected (high risk: 1.00). The prompt explicitly instructs running sudo apt installs, editing system files (e.g. /etc/nginx), configuring services that bind to privileged ports, and starting persistent teamserver/redirector processes—actions that modify the host system state and require elevated privileges.