godmode - Apply prompt-level LLM jailbreaking techniques
Bypasses API-served LLM safety filters using GODMODE, Parseltongue, ULTRAPLINIAN, and persistent Hermes configuration techniques.
Tags
Updated: 2026-10-07Capabilities
Typical Inputs
Typical Outputs
What this skill does
- Apply GODMODE prompt templates
- Obfuscate prompts with Parseltongue
- Race models through OpenRouter
- Configure persistent Hermes jailbreaks
- Auto-detect model families
- Score jailbreak responses
- Undo jailbreak configuration
Inputs
- Target LLM model
- User query
- Hermes configuration
- Jailbreak strategy
- OpenRouter model access
Outputs
- Jailbreak response
- Obfuscated prompt variants
- Model race results
- Updated Hermes configuration
- Prefill message file
- Configuration status report
Requirements
- API-accessible LLM
- Python runtime
- Hermes configuration access
- Configuration write permission
- OpenRouter access for model racing