Clawlet is an ultra-lightweight and efficient personal AI assistant designed for seamless deployment and operation. It addresses the complexity of setting up AI agents by shipping as a single, dependency-free binary, enabling users to run powerful AI capabilities with minimal friction. This makes it ideal for engineers, researchers, and power users seeking a self-contained, highly capable AI agent.

How It Works

Clawlet's core innovation lies in its minimalist architecture. It eschews CGO, runtimes, and external dependencies, bundling SQLite and sqlite-vec directly within the executable. This integrated approach powers a hybrid semantic memory search, allowing agents to efficiently retrieve and utilize past context. The single binary distribution simplifies deployment across diverse environments, ensuring that advanced AI functionalities are readily accessible without complex setup procedures.

Quick Start & Requirements

Installation involves downloading a pre-compiled binary from GitHub Releases. For macOS (Apple Silicon), a typical command is: curl -L https://github.com/mosaxiv/clawlet/releases/latest/download/clawlet_Darwin_arm64.tar.gz | tar xz && mv clawlet ~/.local/bin/. Initial setup requires clawlet onboard to initialize a workspace and configuration. Running Clawlet necessitates API keys for supported LLM providers (OpenAI, OpenRouter, Anthropic, Gemini) or a local LLM setup (Ollama, vLLM). Docker images are also available for containerized deployment.

Highlighted Details

• Features a hybrid semantic memory search, leveraging bundled SQLite and