Repolode
Submit
BaaRai

BaaRai/ARTEX

Open-source AI system that autonomously runs penetration tests using LLM agents, with a Go backend and Next.js web UI.

View on GitHub
Stars
1,659
Forks
625
Open issues
8
Last commit
yesterday

What is it?

ARTEX is an offensive-security platform where LLM-driven agents plan and execute penetration tests against authorized targets. It ships as a Go backend with an embedded Next.js frontend and stores its state in PostgreSQL. Users can track tasks, assets, findings, tool calls and a live exploration graph from a web dashboard.

The core design uses two linked graphs: a globally shared asset graph (domains, subdomains, IPs, services, endpoints) and a per-task exploration graph that records goals, intents, facts and findings. A planner dispatches intents to parallel worker agents, which execute real tools and write results back to both graphs. Workers can also search each other's execution traces, and the planner keeps a shared todolist to advance multi-step attack chains in order.

The project includes a recording MITM proxy for traffic capture, a browser extension for logging into authenticated targets, an approval gate for tool calls, and an asset sync connector for ScopeSentry. It supports Anthropic, OpenAI and local Ollama models, and is licensed under AGPL-3.0 with an explicit restriction to authorized testing only.

Key features

  • 01Autonomous planner and parallel worker agents that run real security tools
  • 02Dual graphs linking a shared asset inventory with per-task exploration lineage
  • 03Traffic recording proxy and browser extension for capturing authenticated sessions
  • 04Approval gate for intercepting and reviewing tool calls before they run
  • 05Vulnerability retesting that records whether a finding is still reproducible or fixed
  • 06Asset sync from ScopeSentry and support for Anthropic, OpenAI and Ollama models

Use cases

  • Running authorized penetration tests on your own systems or a bug bounty program's in-scope targets
  • Lab-based security research into how LLM agents chain multi-step exploits
  • Retesting previously reported vulnerabilities after a fix is deployed
  • Managing discovered assets and findings for a security team in one web console

Quick start

  1. 1

    Clone the repository

    git clone https://github.com/BaaRai/ARTEX.git
  2. 2

    Enter the project directory

    cd ARTEX
  3. 3

    Run the one-click install script, then open http://localhost:8787

    ./install.sh

Things to know

  • Requires a PostgreSQL database.
  • Exploration needs an LLM API key such as ANTHROPIC_API_KEY or OPENAI_API_KEY.
  • Licensed under AGPL-3.0, a copyleft license that requires sharing modified source when distributed or offered as a network service.

In collections

Databases & SQL tools

Topics

#penetration-testing#security-testing#llm-agents#autonomous-agent#golang#nextjs#postgresql#red-team

FAQ

What is ARTEX?
Open-source AI system that autonomously runs penetration tests using LLM agents, with a Go backend and Next.js web UI. ARTEX is an offensive-security platform where LLM-driven agents plan and execute penetration tests against authorized targets. It ships as a Go backend with an embedded Next.js frontend and stores its state in PostgreSQL. Users can track tasks, assets, findings, tool calls and a live exploration graph from a web dashboard.
Is ARTEX free to use? What license does it have?
ARTEX is open source under the AGPL-3.0 license. Check the license terms before using it in commercial products.
Is ARTEX still maintained?
Actively maintained. The last commit to ARTEX was on Nov 6, 2026.
What can I use ARTEX for?
Running authorized penetration tests on your own systems or a bug bounty program's in-scope targets Lab-based security research into how LLM agents chain multi-step exploits Retesting previously reported vulnerabilities after a fix is deployed Managing discovered assets and findings for a security team in one web console
How do I install ARTEX?
Clone the repository: git clone https://github.com/BaaRai/ARTEX.git Enter the project directory: cd ARTEX Run the one-click install script, then open http://localhost:8787: ./install.sh
Where can I get ARTEX?
The source code and installation instructions are on GitHub: https://github.com/BaaRai/ARTEX

Summary generated by AI from the project's README. GitHub stats refresh daily.

Related projects