Fullive-AI/Anima

Make Every Hardware Intelligent — an open-source Agent OS for hardware intelligence

What it solves

Anima is an open-source Agent OS designed to transform passive smart home hardware into proactive, collaborative intelligent agents. It moves beyond simple device control panels by giving hardware the ability to perceive the environment, make decisions based on user preferences, and learn from behavior over time.

How it works

Anima operates as a local network runtime that integrates four core components:

  • Brain: A central decision layer using LangGraph and OpenAI-compatible LLMs to plan actions based on user intent, device state, and memory.
  • Skills: Domain-specific knowledge packages (containing metadata, prompts, and scripts) that define how specific device types (e.g., lights, ACs) should behave and interact.
  • Memory: A layered long-term memory system that extracts preferences and routines from interaction history to evolve device behavior.
  • Adapters: A hardware integration layer that translates structured AI decisions into actual device protocol calls, currently supporting Xiaomi/MIoT devices.

Who it’s for

It is intended for developers and smart-home enthusiasts who want to build an autonomous AI-driven home automation system that learns user habits and supports extensible hardware integration.

Highlights

  • Automatic Device Discovery: Scans local networks and syncs with Xiaomi Cloud via QR login to eliminate manual configuration.
  • Extensible Skill System: Allows users to add custom device intelligence packages without hardcoding logic into the core brain.
  • Evidence-Based Memory: Uses a tiered system (Core Identity, Directory, and Detail) to store and promote confirmed user preferences.
  • Local-First Architecture: Runs on the user's own machine to maintain control over the local network.
  • OpenAI-Compatible: Supports various LLM backends including OpenAI, DeepSeek, and local Ollama endpoints.

Related

  • Project
    omnimind-ai/OmniBot

    An on-device AI agent for Android that integrates chat, system-level tools, and a local workspace to execute complex tasks directly on the device.

  • Project
    AAswordman/Operit

    An open-source AI agent platform for Android that integrates LLMs with system automation, a Linux terminal, and web agents to enable complex task execution on mobile.

  • Project
    the-open-agent/openagent

    An open-source personal AI assistant platform that combines LLMs, RAG, and autonomous agent loops into a single, self-hostable binary.

  • Project
    Sentient-OS-Labs/sentient-os

    An on-device AI operating system for macOS that builds a private, local knowledge base of your digital life to power proactive agents and computer-use automation.

  • Project
    melandlabs/openloomi

    OpenLoomi is an open‑source, cross‑platform desktop “attention agent” that continuously gathers context from your work tools (email, chat, issue trackers, docs) into a searchable, encrypted memory. It surfaces reminders and lets you ask an LLM for help directly inside messaging apps. The runtime is agent‑agnostic, offering Skills and Plugins that integrate with Claude Code, Codex, OpenCode, Hermes, OpenClaw, etc. Installable via simple binaries, it emphasizes local‑first privacy and works out‑of‑the‑box.