Jailbreak Python Packages

jes

Open-source guardrails for AI agents powered by decision models like Jev; checks prompts, retrieved content, tool calls and responses for prompt injection, jailbreaks and secret/PII leaks.

vamp-llm-probe

LLM security auditor — prompt injection, jailbreak, ASCII smuggling detection, bilingual EN+ES red team (900+ adversarial prompts, OWASP Agentic Top 10)

safelabs-eval

Red-teaming and evaluation framework for AI agents, built around an OWASP-inspired ASI01–ASI10 taxonomy

ai-blackteam

Automated LLM red-teaming framework. 1,028 attack techniques, reasoning-layer attacks, and five compliance scorecards (OWASP, MITRE ATLAS, EU AI Act). Vendor-neutral, GPL-3.0.

hermes-jailbench

Zero-LLM deterministic jailbreak regression benchmark: runs a repeatable battery of known-pattern attacks against an LLM endpoint and scores refusal/partial/compliance across runs. Catch when a model or prompt update silently got weaker. Single-turn, responsible-use. pip install hermes-jailbench

humanbound-firewall

Multi-tier firewall for AI agents — blocks prompt injections, jailbreaks, and scope violations. Local tiers first; LLM judge only when uncertain.

prompt-armor

Open-source prompt injection detector — 5 layers, 91.7% F1, ~27ms, offline, Apache 2.0

moabile

A multi-device TUI for mobile app penetration testing (Android & iOS), orchestrating Frida, Objection, device mirroring, and dual-pane file transfer.

llamator

Red Teaming python-framework for testing chatbots and GenAI systems.

sentinel-scan-cli

MCP security scanner (CLI + MCP server): detects tool poisoning, prompt injection, and rug-pulls in MCP manifests. Also an OWASP LLM Top 10 prompt-injection/jailbreak CLI for LLM apps.

basilisk-ai

Basilisk — Open-source AI red teaming framework with genetic prompt evolution. Automated LLM security testing for GPT-4, Claude, Grok, Gemini. OWASP LLM Top 10 coverage. 32 attack modules.

oubliette-dungeon

Standalone AI Red Team Engine - automated adversarial testing for LLM applications

promptlsh

A portable similarity digest for prompt attacks, so reworded jailbreaks correlate instead of reading as unrelated items.

last-layer

Ultra-fast, low latency LLM prompt injection/jailbreak detection ⛓️

bastion-prompt-protection

Lightweight prompt-injection detector built for AI agents, copilots, and LLM apps. Fast, local, calibrated.

agent-security-bench

Open bench for AI coding-agent ML eval + agent security (prompt injection, jailbreak, tool overreach) with machine-readable receipts

protransformers

[NeurIPS 2024] ProTransformer: Robustify Transformers via Plug-and-Play Paradigm. A 4-line robust attention layer that plugs into any pretrained transformer (BERT, ViT, LLaMA, Vicuna, T5, GAT) without retraining.

aiandme

Multi-tier firewall for AI agents — blocks prompt injections, jailbreaks, and scope violations. Local tiers first; LLM judge only when uncertain.

qatration

Adversarial testing for AI chatbots and agents. Runs on your own machine or CI, against your own deployment.

vphone-mcp

📲 One-tool MCP server for app work in a jailbroken vphone iOS VM: IPAs, tweaks, containers, regex-filtered logs

lockllm

Official Python SDK for LockLLM

easyjailbreak

An easy-to-use Python framework to generate adversarial jailbreak prompts.

redteam-foundry

LLM red-team evaluation harness — prompt injection, refusal, leakage and staleness, with cross-judge validation and attack-corpus audits.

semanticshield

The Security Toolkit for managing Generative AI(especially LLMs) and Supervised Learning processes(Learning and Inference).