
cve-2025-32434-poc
Proof-of-concept exploit for CVE-2025-32434, a pickle deserialization vulnerability in PyTorch's weights_only mode, demonstrating malicious tar file…

Proof-of-concept exploit for CVE-2025-32434, a pickle deserialization vulnerability in PyTorch's weights_only mode, demonstrating malicious tar file…

Exploit for CVE-2026-33017, an unauthenticated RCE in Langflow 1.8.1 via the build_public_tmp endpoint, enabling Python code injection through…


vllm-project vllm version 0.5.2.2 is vulnerable to Denial of Service attacks.

Multi-agent LLM system for autonomous penetration testing in isolated labs: planner/worker agents share asset and exploration graphs, with MITM…

Autonomous AI red team framework that chains reconnaissance, exploitation, and post-exploitation into a single pipeline, then triages findings,…

LlamaStack-RCE: Deterministic Supply Chain Exploitation & Hardening Framework [CVE-2024-50050] Focus on AI Security Research…

Autonomous AI pentesting agents — real-time reconnaissance, vulnerability detection, and exploitation orchestration. Go + TypeScript.

Autonomous AI penetration testing agent that orchestrates multi-agent recon, exploitation, post-exploitation, and reporting with persistent…

Python Exploitation Framework, V8 Engine Debugger, Proxy interceptor, marketplace, post-exploitation, backdoor generator,....

Automating Host Exploitation with AI

Android deeplink misconfiguration detector and exploitation tool

Hands-on AI security lab platform with 50+ scenarios across prompt injection, agentic system exploitation, model manipulation, and MCP trust boundary…

Curated collection of research papers on retrieval-augmented generation security, organized by the SLOT taxonomy covering knowledge poisoning,…

A diagnostic framework for measuring LLM vulnerability to Affective Contextual Erosion (ACE) and related liminal attack vectors. **Delirium** is not…

AI-enhanced vulnerability assessment and exploitation toolkit for modular scanning of web applications, detecting SQLi, XSS, RCE, CSRF, and more with…

ExploitGym is a large-scale, realistic benchmark built from real-world vulnerabilities designed to evaluate AI agents' ability to develop exploits.

image scaling attacks for multi-modal prompt injection