
Guardrails
Programmable guardrails for LLM chat apps: enforce input/output rails, block jailbreaks and prompt injections, detect hallucination, and mask…

Programmable guardrails for LLM chat apps: enforce input/output rails, block jailbreaks and prompt injections, detect hallucination, and mask…

Local-first macOS research browser built on a custom Brave build that captures network traffic, fingerprints, scripts, and runtime evidence for…

Code security scanning tool (SAST) to discover, filter and prioritize security and privacy risks.

Security layer for AI coding agents. Works with Claude Code, Cursor, Windsurf, Gemini CLI, OpenCode, Pi Agent and more.

The easiest, most secure way to use WireGuard and 2FA.

Flight recorder and secret scanner for AI coding agents. Reads what Claude Code, Codex, Gemini CLI and 17 more ran, and flags risky actions and…

Peer-to-peer code collaboration and publishing stack with a secure, decentralized protocol, CLI tool, and network daemon for sovereign code forges.

Simulation code and results for a private, verifiable, Byzantine-robust federated intrusion detection protocol using secret sharing, Merkle audits…

IPAnalyzer is an IP Address Tracker OSINT ethical hacking tool built for Linux distributions, designed to gather detailed information about IP…

The Security Toolkit for LLM Interactions

Static and dynamic analysis tool for detecting malicious code, suspicious binaries, and privacy violations

Android client for Prey: reliable device tracking and security tool

A privacy and security engineering toolkit: Discover, understand, pseudonymize, anonymize, encrypt and securely share sensitive and personal data:…

Plugin for Binary Ninja that integrates local Ollama models to rename functions and variables in decompiled HLIL code, preserving privacy by keeping…

The code for ACM MM2024 (Multimodal Unlearnable Examples: Protecting Data against Multimodal Contrastive Learning)

A simple tool to generate fake web browsing and mitigate tracking

A tool that allows you to make TOR your default gateway and send all internet connections under TOR (as transparent proxy) to increase…

Code for the paper Certified Unlearning for Neural Networks, ICML 2025