PicoClaw

PicoClaw is an ultra-lightweight personal AI assistant — a single static Go binary small enough to run on $10 hardware with under 10 MB of RAM. It connects to any OpenAI-compatible LLM (locally via Lemonade or Ollama, or cloud providers such as OpenAI, Anthropic, Gemini, and 30+ others) and to 19+ messaging platforms (Telegram, Discord, WhatsApp, Matrix, and more), performing agentic tasks with persistent memory and skills.

LOCAL AI

Run 'picoclaw.lemonade' to detect a running lemonade-server (http://127.0.0.1:13305), pick one of its models from a menu, and configure it as PicoClaw's default — fully offline, no API key required.

Or run 'picoclaw.inference-snap' to pick a Canonical inference snap (gemma4, gemma3, deepseek-r1, and others) as the provider instead.

SETUP

Run the onboarding wizard on first launch:

picoclaw onboard

Then add a provider API key (or use Lemonade/Ollama locally) and chat:

picoclaw agent -m "Hello!"

The background gateway service (chat integrations) starts automatically the first time you run any picoclaw command and is managed as a standard systemd user unit:

systemctl --user status picoclaw systemctl --user stop picoclaw

SECURITY NOTE

This tool is capable of executing code, accessing files, and interacting with external services on your behalf. Due to the nature of autonomous AI agents, we recommend installing and running this snap inside an LXD container as best practice. A container provides an extra layer of confinement, limiting the agent's reach to the container environment and reducing exposure to your host system and personal data.

lxc launch ubuntu:24.04 my-agent lxc exec my-agent -- snap install --classic picoclaw