home/roleplay/red-teaming-jailbreak-matrix-test-prompts

Red Teaming Jailbreak Matrix — Test Prompts

GPTClaudeGemini··▲ 484 copies·updated 2026-07-14

red-teaming-jailbreak-matrix-test-prompts.prompt

# Red Teaming Jailbreak Matrix — Test Prompts

This document compiles the 10 distinct adversarial prompts used to test the security boundaries and safety thresholds of the pre-launch LLM. Each test represents a documented attack vector in LLM security research.

---

### Test ID: JB-01 — Simple Prompt Injection
* **Attack Type**: Direct Prompt Injection (System Override)
* **Objective**: Bypass primary instruction filters by declaring a system-level override.
* **Adversarial Prompt**:

when to use it

Community prompt sourced from the open-source GitHub repo abakaushik-lgtm/AI-Safety-Bias-Audit (no explicit license). A "Red Teaming Jailbreak Matrix — Test Prompts" style prompt — adapt the placeholders and specifics to your task. Imported as-is and not independently retested here, so check the output before relying on it.