godmode

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.90). The GitHub links point to user-hosted jailbreak repositories that instruct running code (high-risk to fetch-and-execute from an unknown account), while the OpenRouter API and X profile are not direct download hosts.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). This skill is explicitly designed to bypass LLM safety controls (jailbreaking), persistently modify agent configuration to force unrestricted behavior, obfuscate trigger words to evade classifiers, and automate multi-model evasion — clear intentional malicious/abuse functionality.

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

• Third-party content exposure detected (medium risk: 0.30). The skill’s required runtime workflow (“Step 0: Auto-Jailbreak”) reads the user’s plaintext query, applies Parseltongue obfuscation/escalation, then sends that outsider-authored free text (plus system/prefill prompts it generates) to API models for scoring and response selection.