
llm-agent-testbed
An empirical security testbed evaluating prompt injection, confused-deputy vulnerabilities, and tool-calling defenses in LLM agents.

An empirical security testbed evaluating prompt injection, confused-deputy vulnerabilities, and tool-calling defenses in LLM agents.

AI-driven automated threat analysis pipeline that routes files, URLs, IPs, domains, or images through specialized security analyzers and generates…

Zero-trust sandbox for AI agents with kernel-level filesystem jail, transparent network proxy, and YAML-based policy engine to intercept and control…

A secure* runtime for autonomous AI agents. Policy from plain-English constitutions. (*https://ironcurtain.dev)

Policy engine and EDR for AI agent fleets and developer workstations. Monitors tool calls, file access, network flows, and process execution with…

OS-level monitor for AI agents: observes processes, file access, and network activity on the local machine and attributes each event to an agent…

Read-only WordPress security scanner for HestiaCP servers. Detects wp2shell compromise indicators (CVE-2026-63030 / CVE-2026-60137) across all hosted…

Host-agnostic pre-write security hook for coding agent: detects user-input patterns via Semgrep and emits deterministic, no-LLM security guidance.

Docker Container Escape POC via mlx-metal importlib

Automated steganography detection tool that scans websites, web servers, and local directories using AI-driven object/text recognition and deep file…

AI Prompt Secret Scanner: local proxy and Claude Code hook that blocks secrets before they reach AI APIs

Multi-layer security framework for AI agent ecosystems. Provides pre-installation skill auditing, file integrity monitoring, runtime protection, and…

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

Lightweight, cross-platform process sandboxing powered by OpenAI Codex's runtime. Sandbox any command with file, network, and credential controls.

Alignment-research scaffold (autoresearch-style) for LLM guardrails: search over a single policy.md surface