Skip to content
Hooman

Getting Started

Hooman is the full-stack open-source agentic ecosystem: CLI, VS Code, ACP, daemon channels, and Design mode on one local-first runtime. Built with TypeScript, the Strands Agents SDK, and Ink.

Everything — configuration, API keys, sessions, skills — lives in a single ~/.hooman directory shared across every surface. No account. No telemetry. MIT licensed — BYOK, custom inference endpoints, or fully offline with llama.cpp / MLX.

Pick your path

Section titled “Pick your path”
Terminal / CLI →

curl | bash, npx, or bunx — no clone, no build.

 VS Code →

A native chat panel in the activity bar, from the Marketplace.

Requirements

Section titled “Requirements”
  • Node.js >= 24 or Bun >= 1.1
  • Credentials or a local runtime for whichever LLM provider you choose — see Models

By default Hooman’s first-run setup wizard offers local llama.cpp presets for Gemma 4 E2B and Qwen3.5 2B on Q4_K_M, plus — on Apple Silicon — local MLX presets for Gemma 4 E2B and Qwen3.5 2B. You can finish setup with no API keys; the first local model you use downloads from the Hugging Face Hub on demand with live progress (percent, size, speed, ETA) shown in every surface. Swap in a hosted provider during setup or later via hooman config / Settings whenever you’re ready.

What’s next

Section titled “What’s next”
  • CLI quickstart — install, first-run setup, and your first hooman chat session.
  • VS Code quickstart — install the extension and complete setup in the activity bar.
  • CLI reference — chat, setup, exec, daemon, and /config in depth.
  • Modes — Agent, Plan, Ask, and Design.
  • Design mode — HTML artifacts, DESIGN.md, preview, and export to PDF, PowerPoint-ready .pptx, Figma-ready .fig / .deck, or Sketch-ready .sketch.
  • Configuration — ~/.hooman layout and repo-local overlays.
  • MCP — tool servers, OAuth, and event-driven automation.