Automated Red Teaming Engine with Multi-agent Intelligent Supervision
ARTEMIS is an autonomous agent created by the Stanford Trinity project to automate vulnerability discovery.
Install uv if you haven't already:
curl -LsSf https://astral.sh/uv/install.sh | shInstall the latest version of Rust (required for building):
# Remove old Rust if installed via apt sudo apt remove rustc cargo sudo apt install libssl-dev # Install rustup (the official Rust toolchain installer) curl --proto '=https' --tlsv1.2 -sSf https://sh.rustup.rs | sh # Restart shell or source the environment source ~/.cargo/env # Install latest stable Rust rustup install stable rustup default stable
First, we have to build the codex binary:
cargo build --release --manifest-path codex-rs/Cargo.toml
Now we can setup the Python environment:
uv sync
source .venv/bin/activateCopy the example configuration and add your API keys:
cp .env.example .env
# Edit .env with your API keysRequired environment variables:
OPENROUTER_API_KEYorOPENAI_API_KEY- For the supervisor and LLM callsSUBAGENT_MODEL- Model to use for spawned Codex instances (e.g.,anthropic/claude-sonnet-4)
Try a simple CTF challenge to verify everything works:
python -m supervisor.supervisor \ --config-file configs/tests/ctf_easy.yaml \ --benchmark-mode \ --duration 10 \ --skip-todos
This runs a 10-minute test on an easy CTF challenge in benchmark mode (no triage process).
For detailed configuration options and usage, see supervisor-usage.md.
Build the Docker image:
docker build -t artemis .Same as above - copy the example configuration and add your API keys:
cp .env.example .env
# Edit .env with your API keysRequired environment variables:
OPENROUTER_API_KEYorOPENAI_API_KEY- For the supervisor and LLM callsSUBAGENT_MODEL- Model to use for spawned Codex instances (e.g.,anthropic/claude-sonnet-4)
If using OpenRouter, you'll need to configure the codex binary. Create ~/.codex/config.toml:
mkdir -p ~/.codex cat > ~/.codex/config.toml <<'EOF' model_provider = "openrouter" [model_providers.openrouter] name = "OpenRouter" base_url = "https://openrouter.ai/api/v1" env_key = "OPENROUTER_API_KEY" [sandbox] mode = "workspace-write" network_access = true EOF
Use the provided run_docker.sh script:
# Run with OpenRouter (mounts ~/.codex/config.toml) ./run_docker.sh openrouter # Run with OpenAI only (no config mount needed) ./run_docker.sh openai
The script will:
- Mount your
~/.codex/config.toml(if using OpenRouter) - Mount the
./logsdirectory for persistent logs - Use your
.envfile for API keys - Run a 10-minute test on an easy CTF challenge
Manual Docker Run:
If you prefer to run docker manually:
# With OpenRouter docker run -it \ --env-file .env \ -v $HOME/.codex/config.toml:/root/.codex/config.toml:ro \ -v $(pwd)/logs:/app/trinity/ARTEMIS/logs \ artemis \ python -m supervisor.supervisor \ --config-file configs/tests/ctf_easy.yaml \ --benchmark-mode \ --duration 10 \ --skip-todos # With OpenAI only docker run -it \ --env-file .env \ -v $(pwd)/logs:/app/trinity/ARTEMIS/logs \ artemis \ python -m supervisor.supervisor \ --config-file configs/tests/ctf_easy.yaml \ --benchmark-mode \ --duration 10 \ --skip-todos
This project uses OpenAI Codex as a base, forked from commit c221eab.
This repository is licensed under the Apache-2.0 License.