One binary.
Sixty seconds.
Prebuilt binaries for Linux and macOS, x86_64 and aarch64 — and on Windows, inside WSL2. The install script verifies a SHA-256 checksum before installing anywhere on your PATH.
Prebuilt binary
Recommended path. No Rust toolchain required.
Cargo
If you'd rather compile. Requires Rust 1.82+.
From source
The release binary lands at target/release/axocoatl.
Windows (WSL2)
Axocoatl is Unix-first — the session sandbox is Podman, the service is
systemd/launchd — so on Windows it runs inside WSL2, not natively. Set up
WSL2 (wsl --install in an admin PowerShell), then inside your
distro install Podman and run the same one-liner:
Run axocoatl dev in WSL, then open http://localhost:8080 in
your Windows browser. Full steps in the
Windows guide.
After install
axocoatl doctor: verify Ollama, podman, and the data dir are reachable.axocoatl onboard: interactive setup wizard. Picks a provider, scaffolds a project.axocoatl dev: start the daemon, IPC, and dashboard on:8080.- Open
http://localhost:8080. Switch to Sessions. Click the seeded demo-counters session.
Bring your own LLM
Axocoatl runs against Ollama (local), OpenAI, Anthropic,
Mistral, Gemini, and OpenRouter. Pick one in onboard
or set it manually in axocoatl.yaml. Each agent
picks its provider independently: local model for one, cloud
for another, same workflow.