#1LLM security, prompt injection, model extraction, adversarial AI, and AI red teaming tools.
Kitploit recommended

Alignment-research scaffold (autoresearch-style) for LLM guardrails: search over a single policy.md surface
Lightweight, secure Linux sandboxes for untrusted processes. Runs in the browser and on the server.

Research code for poisoning attacks on the PGM-index, demonstrating how to craft adversarial data to degrade learned index performance.

Black-box attack framework that hijacks reasoning in agentic retrieval-augmented generation systems by injecting poisoned documents, with support for…

a guard that blocks catastrophic agent actions

Powerful protection for AI agents - Open-source security and cost tracking for AI applications

Professional vulnerability assessment report for LiteLLM command injection risk, including executive summary, technical impact, remediation, and…

Reproduction code for a controlled study of membership-inference attacks on OLMo 2, covering document insertion, mid-training, post-training, scoring…

Train, evaluate, and explore neural networks with built-in adversarial robustness tools, including PGD attacks, adversarial training, and robust…

Fingerprint OpenAI-compatible LLMs from tokenizer and behavior signals.

Zero shot vulnerability discovery using LLMs


Research code for adding controllable safety experts to LLM layers via two-stage upcycling training, with inference-time safety temperature routing…

Weight-level LLM forensics: diff model weights, detect fine-tunes, merges and disguised derivatives, and surface tokens a fine-tune promotes or…

Code and paper source for a NeurIPS EvoRobust study showing simple bag-of-words and metadata features beat most AI control monitors at detecting…

Open-source sandboxed agent harness for teams. Giving every employee a secured personal agent.

Trap AI web scrapers in an endless poison pit.

Purple Team Exercise Framework