ARTEX is an autonomous penetration testing system that combines an AI agent with a Go backend and Next.js frontend to conduct security assessments.
The system addresses the challenge of automating penetration testing by deploying an AI agent that can independently explore target systems, discover vulnerabilities, and document findings. The agent operates through a task-driven workflow where it executes reconnaissance and exploitation steps, records its actions in session logs, and maintains a graph of discovered assets and relationships. Users can configure LLM providers via environment variables or the UI, and the system supports integration with ScopeSentry for importing existing asset inventories to seed the agent's exploration scope.
Organizations should adopt this tool if they want to automate initial reconnaissance and vulnerability discovery phases while maintaining human oversight. The system suits security teams conducting authorized penetration tests who need visibility into agent decision-making and the ability to approve or block actions before execution. The tool includes traffic recording, human-in-the-loop dialogue capabilities, and an asset coverage visualization that highlights tested versus untested targets. Installation can proceed either through a one-command script that handles Docker setup or local compilation, with PostgreSQL as the required database backend.
Development activity shows consistent maintenance of the codebase with regular updates addressing user-reported issues and feature requests. The project maintains active documentation including screenshots and deployment guides that reflect current functionality. Contributors respond to configuration and installation problems reported by users, and the team has transitioned Docker image hosting to ensure continued availability of pre-built artifacts.
