ZeroClaw

What is ZeroClaw?

ZeroClaw is an open-source personal AI agent: a single Rust binary that you run on your own hardware, with your own API keys or local models. It talks to 70+ LLM providers and reaches 30+ channels, including Telegram, Discord, Slack, Signal, Matrix and email, so you can message an agent that lives on your workstation, homelab or a Raspberry Pi. Safety is a core design point: supervised autonomy is the default, medium-risk actions wait for approval, high-risk ones are refused, and sandboxes such as Landlock, Bubblewrap, Seatbelt or Docker limit what a process can touch. Optional tool receipts add an HMAC-SHA256 tag to each successful tool call so a claimed action can be checked. Installation is a one-line script, Homebrew or Docker, and the project is licensed MIT or Apache-2.0. There is no hosted version, account system or subscription.

Use cases

• Chat with a self-hosted assistant through Telegram
• Run shell automation that asks before risky commands
• Schedule agent jobs on a homelab
• Use local models with Ollama or llama.cpp
• Run many isolated named agents from one daemon
• Control Raspberry Pi or Arduino hardware from an agent
• Verify tool calls with signed receipts

Who it's for

• Self-hosters
• Developers
• Homelab users
• Security-minded teams

Why use ZeroClaw

• The agent runs on your own hardware, so a local model keeps prompts off third-party servers.
• Approval gates and sandboxes mean the agent does not get a blank cheque to run commands.
• It fits on small machines, so one setup can span a workstation, homelab and single-board computer.
• Receipts let you detect a tool run the model claims but never made.
• No subscription or cloud seat is needed; the cost is the model usage you choose.

ZeroClaw features

• Single lightweight Rust binary with a prebuilt installer
• 70+ LLM providers plus local runtimes such as Ollama, LM Studio, llama.cpp and vLLM
• 30+ channels including Telegram, Discord, Slack, Signal, Matrix and email
• Supervised autonomy with approval prompts for risky actions
• Sandboxing through Landlock, Bubblewrap, Seatbelt or Docker
• HMAC-SHA256 tool receipts the model cannot forge
• Multi-agent runtime with per-agent tool allowlists
• Skills, WASM plugins and A2A agent discovery
• Cost tracking ledger with daily and monthly budgets
• Runs on Linux, macOS, Windows, FreeBSD, NixOS and Raspberry Pi class boards

ZeroClaw integrations

• Telegram
• Discord
• Slack
• Signal
• Matrix
• WhatsApp
• Mattermost
• iMessage
• Ollama
• LM Studio
• llama.cpp
• vLLM

How to get started with ZeroClaw

1. Install the binary with the one-line script, Homebrew or Docker.
2. Run zeroclaw quickstart, choose a model provider (a local Ollama install or a key of your own) and link a messaging channel.
3. Start agent mode for interactive chat or one-shot prompts with tools and memory.
4. Run the gateway and daemon so channels, webhooks and schedules can trigger the agent.
5. Check status and doctor output, and adjust autonomy and allowlists as needed.

ZeroClaw limitations

• There is no hosted version, account system or billing; you run and maintain it yourself
• It is an agent runtime you configure, not a framework for building apps on top
• Model costs are paid to the providers whose keys you use
• The project is on the v0.8 series, with a stable 1.0 still planned

ZeroClaw pricing

ZeroClaw is open source, so you can use and self-host it for free.

ZeroClaw plans

• Open source

  Free

  self-hosted, no subscription

  • Full runtime on your own machine
  • Use your own keys or local models
  • MIT or Apache-2.0 license

Plans as listed on the official pricing page on 24 Oct 2026. Prices can change and may vary by country or exclude tax. Check the official pricing page.

ZeroClaw alternatives

The main ZeroClaw alternatives listed on Debuly are OpenClaw, Hermes Agent, Open Interpreter, goose, AutoGPT and Moltworker. The table compares pricing and platforms; order is A–Z.

1. 

   AutoGPT

   Agent platform with a plain-English AutoPilot, a visual workflow builder and self-hosting, plus a $50/month Pro plan with a 7-day trial.

   Automation and agentsDeveloper platformsProductivity
   FreemiumVisit AutoGPT (official site)
2. 

   goose

   Open source AI agent that runs on your machine as a desktop app, CLI or API and connects to tools through MCP extensions.

   Automation and agentsCodingDeveloper platforms
   Open sourceVisit goose (official site)
3. 

   Hermes Agent

   Free, MIT-licensed AI agent from Nous Research with persistent memory, auto-built skills, scheduling and chat-app access.

   Automation and agentsCodingProductivity
   Open sourceVisit Hermes Agent (official site)
4. 

   Moltworker

   Cloudflare's proof of concept for self-hosting the OpenClaw personal AI agent on Workers and Sandboxes instead of a Mac mini.

   Automation and agentsDeveloper platformsChat assistants
   Open sourceVisit Moltworker (official site)
5. 

   Open Interpreter

   Open-source terminal coding agent, built in Rust on Codex, that emulates the agent harness suited to each low-cost model.

   CodingAutomation and agentsDeveloper platforms
   Open sourceVisit Open Interpreter (official site)
6. 

   OpenClaw

   Open-source personal AI assistant that runs on your own devices and takes actions for you through WhatsApp, Telegram, Slack, Discord and other chat apps.

   Automation and agentsChat assistantsProductivity
   Open sourceVisit OpenClaw (official site)

ZeroClaw FAQ

Sources & verification

Facts on this page come from the official sources below, each with the date we read it. Written by Debuly; the vendor did not review this page.

Image credits

• Logo: from zeroclaw.com · fetched 24 Oct 2026 · Vendor trademark, used only to identify the product; removed on request
• Cover: none collected yet, so a generated tile is shown.

Spot an error? Report an issue with this listing. We also remove a logo or image on the owner's request.