A multi-Agent collaboration platform for authorized penetration testing and vulnerability discovery
A lead Agent coordinates specialist Agents for intelligence gathering, validation, code audit, reverse analysis, and cryptanalysis.
WorkProject binds graph-targeted WorkItems to authorized assets, WorkItem-attributed evidence, validated findings, continuous attack paths, retest candidates, and lead review decisions.
Building knowledge graphs with LightRAG Core provides matching original document chunks and graph context for task-oriented inputs.
The UI consumes normalized timeline events that can be streamed live or loaded later as history.
Managed Docker hosts, images, and containers allow execution environments to be isolated, scaled, and assigned to projects.
The default sandbox image bundles recon, DNS, web discovery, credential testing, Android, firmware, reverse engineering, browser, Python, and wordlist capabilities behind sandbox-local skills.
Container traffic can be routed through direct, HTTP, HTTPS, or SOCKS5 modes using one platform-managed policy surface.
The frontend combines chat, workflow state, graph review, evidence chains, attack paths, sandbox selection, terminal, files, and noVNC.