autonomous-ai

Autonomous Lamp — AI skill for Claude Code

AI community

Archived — moved to github.com/autonomous-ai/autonomous-os.

How to install Autonomous Lamp

This entry records only its repository, not the path inside it, so there is no exact command to give. Open autonomous-ai/autonomous-lamp and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Autonomous Lamp does

Archived — moved to github.com/autonomous-ai/autonomous-os.

Alternatives in AI

Ralph For Claude Code — by Frank Bria - An autonomous AI development framework that enables Claude Code to work iteratively on project 8k ★
Osaurus — Own your AI. The native macOS harness for AI agents -- any model, persistent memory, autonomous execution, cry 5.1k ★
Ralphex — Extended Ralph loop for autonomous AI-driven plan execution 1.1k ★

README

[!IMPORTANT]

This repository has moved. Active development continues at [autonomous-os](https://github.com/autonomous-ai/autonomous-os).

Please update your bookmarks, links, and Git remote URLs. This repository is retained for historical reference only.

Autonomous Lamp

https://github.com/user-attachments/assets/2e6eea7d-312e-47dd-94cf-0914dedeccc4

Autonomous Lamp is an AI-powered lamp built on a Raspberry Pi or OrangePi. It listens, sees, moves, and talks. The brain is a pluggable agentic gateway (OpenClaw, Hermes, or any other LLM + skills + memory runtime); the hands are LeLamp (servo, LED, mic, speaker, camera, display); Lamp Server (Go) glues them together with networking, sensing, OTA, and a web UI.

Target hardware: Raspberry Pi 4/5 or OrangePi (any Linux ARM64 SBC with comparable I/O).

Gallery

Architecture

Three layers, each independently replaceable:

Agentic Gateway (LLM / skills / memory)        ← OpenClaw, Hermes, or any agentic runtime
        │  WebSocket
        ▼
Lamp Server (Go, :5000)                        ← system, network, MQTT, OTA, sensing routing, local intent
        │  HTTP
        ▼
LeLamp Runtime (Python, :5001)                 ← hardware drivers (servo, LED, audio, camera, display)
        │
        ▼
Hardware (plug-and-play; missing parts are skipped)

Design principles:

Hardware is a plugin. If a device is missing, that subsystem is skipped — the rest still runs.
The brain is swappable. Lamp treats the agentic gateway as an abstract dependency — OpenClaw today, Hermes or another runtime tomorrow. The interface is a WebSocket plus a SKILL.md contract.
System layer runs without the gateway. Network, OTA, reset, LED feedback, local intents all work offline.
LeLamp has no AI. Drivers o