Waku Agent

Waku Agent is a personal AI assistant that runs on your laptop, stores its memory locally and calls remote model providers. It's for developers who want to inspect and change the assistant's behavior in Python. The agent uses provider API credentials for reasoning; requests to cloud providers are processed by those services.

Supported providers include Anthropic, OpenAI, Gemini, DeepSeek, MiniMax, Kimi and GLM, alongside OpenRouter, OpenCode Zen and OpenCode Go. Requests go to the selected provider, while the assistant's persistent memory lives in a SQLite file on your machine.

That memory covers facts, past interactions and reusable skills. The code includes rules for deciding what to remember and what to retain over time. Developers can change those rules, the agent loop and the evaluation code rather than relying on a fixed assistant product.

You can talk to the assistant through a terminal, local browser dashboard or voice, with gateways for Telegram, Discord and WhatsApp. The dashboard displays messages during processing. Per-turn traces, deterministic tests and model-based evaluations help developers assess changes to its behavior.

Optional Waku Memory requires sign-in and stores shared memory remotely for use across Claude Code, Codex, Hermes and Grok Bot. The agent code is open source under MIT. The separate hosted platform code uses Elastic License 2.0, which permits self-hosting and commercial use but prohibits offering it as a service to others.