codex-redteam-mode
Fail
Audited by Snyk on May 20, 2026
Risk Level: CRITICAL
Full Analysis
CRITICAL E005: Suspicious download URL detected in skill instructions.
- Suspicious download URL detected (high risk: 0.90). Suspicious — the set includes an unknown GitHub repo that instructs running installer scripts (including PowerShell with ExecutionPolicy Bypass) and a non-standard domain (ara.so) used by the skill, which together are common vectors for distributing malicious tooling (the example target domains are placeholders but the installer instructions amplify risk).
CRITICAL E006: Malicious code pattern detected in skill scripts.
- Malicious code pattern detected (high risk: 0.90). High-risk dual-use tooling: although framed as opt-in and for authorized assessments, the skill explicitly enables creation of reverse shells/C2 payloads, AV/EDR evasion, persistence, privilege-escalation/AD exploitation, and installer behavior that injects code into existing configs — all clear patterns that facilitate backdoors, remote code execution, credential theft, data exfiltration, and supply-chain persistence.
MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).
- Potentially malicious external URL detected (high risk: 0.90). The installation explicitly runs "git clone https://github.com/chAng-L19/codex-redteam-mode.git" followed by executing scripts/install.py, which fetches and executes remote repository code that injects hooks and runtime prompt overlays that directly control agent prompts, so this URL is a required runtime dependency that executes remote code.
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 enables red-team/post-exploitation workflows (payload generation, AV evasion, persistence), includes a PowerShell "ExecutionPolicy Bypass" installer and instructions that modify runtime/hooks files and system state, which direct the agent to bypass security controls and alter the host for persistence—so it encourages compromising the machine.
Issues (4)
E005
CRITICALSuspicious download URL detected in skill instructions.
E006
CRITICALMalicious code pattern detected in skill scripts.
W012
MEDIUMUnverifiable external dependency detected (runtime URL that controls agent).
W013
MEDIUMAttempt to modify system services in skill instructions.
Audit Metadata