NanoClaw

NanoClaw is a lightweight, secure personal AI assistant that runs each agent in its own isolated container on your own server.

  • AI
  • Automation
  • Other
  • Tools & Utilities
NanoClaw is a lightweight, secure personal AI assistant that runs each agent in its own isolated container on your own server. It provisions agents (for example in Slack) so every one gets its own bot identity, container, and memory, and lets you spawn specialised teammate agents from chat. Built as a minimal, fully readable alternative to OpenClaw, it's aimed at self-hosters who want a controllable AI agent they can actually understand and customise.

Platform availability

Available on 2 of 10 platforms

  • Unraid
  • TrueNAS (not listed)
  • Umbrel (not listed)
  • ZimaOS (not listed)
  • Proxmox (not listed)
  • Helm (not listed)
  • Mac (not listed)
  • Windows (not listed)
  • Linux (not listed)
  • Any server

Health score?A 0-100 score computed nightly from four factors: maintained (40%: recent commits and steady activity), popular (25%: GitHub stars, log scale), easy to install (20%: platforms, install notes, setup guides, Docker image) and light to run (15%: minimum RAM, ARM support). When the source code has been checked, a fifth factor, security check (20%: automated scan for committed secrets, unsafe desktop-app settings, known-vulnerable dependencies and project hygiene; any serious finding caps it at 20), is added. Factors without data are left out and the rest are rescaled.

75/100 Good

Maintained
100last commit 26 days ago
Popular
9031,059 GitHub stars
Easy to install
452 install platforms, install notes, Docker image
Light to run
304 GB minimum RAM
Security check
70no serious findings, 8 minor, CI, tests

Availability: platform listings confirmed today, links checked 55 days ago

GitHub stars
31,059
Open issues
1033
Last commit
2026-10-26
Latest release
v2026.10.0
License
MIT
Activity
declining

Checked 22 days ago - source: GitHub (qwibitai/nanoclaw)

Resources & Compatibility

Minimum requirements: 4 GB RAM. ARM (e.g. Raspberry Pi) is not supported.

Checked 54 days ago - source: Docker Hub image tags and published docs

First-install notes

  • Run the installer script Clone the repository and run `bash nanoclaw.sh`; it installs missing basics like Node (JavaScript runtime), pnpm (package installer), and Docker (container runtime), and registers Anthropic credentials for Claude. If you skip this, the app may fail to start because required tools or the Claude API login are missing. source
  • Start messages with @Andy Send messages that begin with the trigger word, for example `@Andy help me`. If the trigger is missing or comes later in the sentence, NanoClaw ignores the message. source
  • Check the macOS service status On macOS, use launchctl (macOS's background-service tool) to start, stop, or check NanoClaw's background service. If the service isn't running, the agent won't respond even if the install looks complete. source

Fetched 54 days ago - each note links to its source

Alternatives (12)

  • Alpaca - Ollama Client - Alpaca is an Ollama client where you can manage and chat with multiple models, Alpaca provides an easy and begginer friendly way of interac…
  • AnythingLLM - AnythingLLM is a local, all-in-one AI app for self-hosters that works with any large language model, providing document chat, retrieval-aug…
  • ChatGPT - ChatGPT is a generative artificial intelligence chatbot developed by OpenAI and launched in 2022.
  • Claude - AI assistant for chat and API, performing summarization, search, writing, Q&A, and coding with high reliability, customizability, priva…
  • DeepSeek - DeepSeek is a state-of-the-art AI assistant designed to provide intelligent, conversational support across diverse tasks.
  • Ensu - Private offline on-device local LLM app by Ente, open source code, no accounts or tracking, and works cross-platform on mobile and desktop…
  • Hermes Agent - Hermes Agent is a self-improving AI agent by Nous Research that creates and refines skills from experience, builds a persistent model of yo…
  • Jan.ai - Open-source offline AI chat software supporting local LLMs, cloud model connections, custom assistant creation, OpenAI-compatible API, and…
  • Microsoft Copilot - Microsoft Copilot, formerly known as Bing Chat, is an AI chat interface used to converse with Microsoft hosted OpenAI and Claude models wit…
  • OpenClaw - OpenClaw is an open-source, self-hosted AI assistant gateway that connects large language models (LLMs) to Discord, Telegram, and WhatsApp,…
  • Poe - Poe is a web-based AI platform that hosts multiple assistant bots—including Claude and various GPT and image/audio models—to provide conver…
  • Secret Llama - Fully private offline chatbot runs locally in browser, with no server or installation needed.

Checked 51 days ago - source: AlternativeTo

Common questions

How do I install NanoClaw on a new server?

Clone the repository and run bash nanoclaw.sh from a fresh checkout. The script installs missing Node, pnpm, and Docker, registers an Anthropic credential, builds the agent container, and pairs your first channel. If a step fails, Claude Code is invoked automatically to diagnose and resume.

How do I migrate an existing NanoClaw v1 install to v2?

Run bash migrate-v2.sh from a fresh v2 checkout next to v1, not from inside a Claude session. It merges the .env file, seeds the v2 database from registered groups, copies group folders, session data, scheduled tasks, selected channel adapters, and channel auth state, then builds the container. It does not flip the service, so switch to v2 only after testing.

Why does the agent ignore some of my messages?

Messages must start with the configured trigger word, such as @Andy help me. If the trigger is missing or appears later in the sentence, NanoClaw ignores the message.

What should I check if a v2 setup stops working or drops messages?

Check for host startup stops, container spawn errors, dropped messages, silent agents, and stuck sessions.

Answers sourced from docs.nanoclaw.dev, github.com, github.com

Community

No community discussions collected yet.