
agent-trace-detector-eval
Code, derived data and results for: Evaluating Baseline Detectors of Hijacked LLM Agents Under Rare Attacks and Distribution Shift

Code, derived data and results for: Evaluating Baseline Detectors of Hijacked LLM Agents Under Rare Attacks and Distribution Shift

Volatility 3 plugin that carves MCP/JSON-RPC message candidates from Windows and Linux memory dumps, preserving raw bytes, addresses, and SHA-256 for…

Self-hosted multi-engine Sigma detection for event logs

User-space bot and DDoS deflection for small HTTP services

Free access-log compromise analyzer: webshell interactions, exploitation patterns and brute-force bursts in your Apache/nginx logs - severity,…

A single-file native Windows tray application (C, Win32, no runtime) that shows exactly which program is sending and receiving bytes and to which…

Open-source, self-hosted security pipeline for IT, AI and OT. Logs in. Evidence out.: OCSF normalization, 46 detection rules…

Reverse-proxy web application firewall with WAF, anti-bot and anti-DDoS protection at L3/L4/L7, custom rules, rate limiting, geo/IP filtering and…

Hunt WiFi threats silently. Catch the hackers.

🐍 - Crawler on tor network sites, searching for keywords.

Streamlit educational dashboard that simulates port scanning, DoS, brute force and data exfiltration attacks, with visualizations, attack logs and…

Unsupervised deep autoencoder that flags anomalous mobile traffic: 25,000+ HTTP/HTTPS flows intercepted from a live iOS device with mitmproxy,…

Installs and configures a DShield honeypot sensor on Raspberry Pi or Ubuntu, running Cowrie SSH and HTTP honeypots and forwarding firewall logs to…

High-interaction OPC UA honeypot for ICS/SCADA — multi-profile (oil & gas, water, power grid), live process simulation, per-session threat scoring,…

Read-only triage script for WordPress sites after PHP2Shell (CVE-2026-109618, CVE-2026-109619). Checks for signs of compromise and writes a report…

Sentinel — DeanWay Labs' inbound governor. Guards what an AI agent hears: world content is data, never orders. The mirror of QuietOS.

OWASP reference project for agentic AI security evidence: correlates agent execution, identity, tool calls, and authorization decisions into…

An audit-oriented risk model for how agent-generated inferences can change epistemic status, gain operational authority, and influence later…