codex

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 0.90). This skill documentation dangerously encourages autonomous, unrestricted code execution (--full-auto / danger-full-access / skip-git-repo-check) that enables remote code execution, broad filesystem and network access and thus presents a high risk of data exfiltration and system compromise if misused.

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 promotes modes like "Danger-Full-Access" that provide network- and system-level, all-files access and encourages autonomous, full-auto file editing without requiring approval, which directly enables modifications to the machine state (including actions that would require sudo or affect system files).