# LLM Jailbreak Multimodal

> Multimodal jailbreaks: typography-in-image, adversarial pixels, screenshot policy text, PDF polyglots, audio stego.

- Skill: `netvar1337/llm-jailbreak-multimodal` (Agent Skill)
- Install (CLI): `npx skillmds@latest add netvar1337/llm-jailbreak-multimodal`
- Raw SKILL.md: https://api.skillmd.com/api/skills/netvar1337/llm-jailbreak-multimodal/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Design & Media
- License: GPL-3.0-or-later
- Author: netvar1337 (https://skillmd.com/u/netvar1337)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/netvar1337/llm-jailbreak-multimodal

---

# Multimodal jailbreaks

## Channels
- Vision-language: render instructions as image text / screenshots
- Adversarial perturbations (research stacks with white-box access)
- Documents: PDF with hidden text, tiny font, white text
- Audio: TTS instructions, spectrogram stego (if supported)
- Video frames / OCR pipelines

## Practical playbook
1. Confirm modality path actually reaches the model (not stripped).
2. Typography attack first (high reliability).
3. Combine image instruction + short benign user text.
4. For indirect: host image/PDF where RAG/fetch will ingest.

## Evaluation
OCR the image yourself; if you can't read it, model may not either.