KernDocs
KernGitHub ↗

Local deployment

Self-host on macOS or Linux

Deploy Kern locally in a Lima virtual machine and open the admin UI securely.

Requirements

  • macOS or Linux with hardware virtualization enabled
  • Lima 1.x or 2.x installed and available as limactl. On macOS, run brew install lima.
  • Python 3.11 or newer
  • At least 2 CPUs, 2 GiB RAM, and 48 GiB disk available to Lima

Deploy the host

Clone and prepare credentials
git clone https://github.com/infiloop2/kern.git
cd kern
python3 -m host.cli.generate_password
ssh-keygen -t ed25519
Deploy with Lima
python3 -m host.cli.deploy \
  --provider lima \
  --agent-name my-kern \
  --admin-password-sha256 <sha256-from-generate-password> \
  --operator-ssh-public-key "$(cat ~/.ssh/id_ed25519.pub)"

Open Kern over SSH

bash
ssh -p <ssh-port> -N -L 7443:127.0.0.1:7443 [email protected]

Keep the tunnel running, open http://127.0.0.1:7443, and sign in with the admin password you saved.

Optional: add remote browser access

When you want a stable HTTPS address that works from any browser, follow Set up remote access to add a Cloudflare Tunnel.

From deployment to a useful first run

  1. 1
    Deploy on your computer

    Create the Lima host, open the SSH tunnel, then sign in to Kern on localhost.

  2. 2
    Connect an agent

    Open Home → Integrations and connect the runtime you use.

  3. 3
    Enable only the access you need

    Connect tools and network integrations for the task. Kern denies everything else.

  4. 4
    Give the agent a real outcome

    Start a Chat with a bounded task, such as researching a question or improving a repository.

  5. 5
    Inspect the boundary

    Review Network audit and Tool audit to see what was allowed, denied, or held for approval.