#1LLM security, prompt injection, model extraction, adversarial AI, and AI red teaming tools.
Kitploit recommended
michaels1011/ephemora-cellCapability-based WASM runtime for executing untrusted AI-generated code with enforced CPU, memory, time, I/O, and filesystem limits. Provides…
nanda-rani/safeA contextual security auditing system for research artifacts
db-ol/cuaharmBenchmark for evaluating safety risks of computer-using agents, with 104 realistic misuse scenarios across seven malicious categories, supporting…
ghostwolflab/ai_security_topA structured knowledge base covering AI security fundamentals, threat modeling, red team offensive techniques, and blue team defenses, including LLM…
emrekybs/signalopsAll-source intelligence fusion dashboard — WiFi, cellular, CCTV, ADS-B, orbital & SDR feeds unified into a single tactical map. Security research…
atsukisato/pgm-attackResearch code for poisoning attacks on the PGM-index, demonstrating how to craft adversarial data to degrade learned index performance.
cucu220123/safety-awarenessResearch code for extracting and training safety-awareness directions in multimodal LLMs to improve refusal behavior while limiting benign-task drift.
qkrrkd90-source/weavemarkReference implementation of a multi-bit LLM watermarking scheme using coded payload spreading, unbiased reweighting, and soft-decision ECC decoding…
vxssroott/cve-2026-78906-chatgpt-prompt-injectionAI Infrastructure Vulnerability Research. CVE-2026-78906: Prompt injection and memory exfiltration in OpenAI's ChatGPT API.
trailofbits/privacyravenPrivacy testing library for deep learning systems, enabling assessment of susceptibility to membership inference, model extraction, and model…
protectai/rebuffMulti-layered prompt injection detector for AI applications using heuristics, LLM-based analysis, vectorDB attack signatures, and canary token leak…
protectai/llm-guardThe Security Toolkit for LLM Interactions
doubangotelecom/facelivenessdetection-sdk3D Passive Face Liveness Detection (Anti-Spoofing) & Deepfake detection. A single image is needed to compute liveness score. 99,67% accuracy on our…
