# zeroclaw-labs/zeroclaw

Fast, small, and fully autonomous AI personal assistant infrastructure, any OS, any platform — deploy anywhere, swap anything 🦀

Repository: https://github.com/zeroclaw-labs/zeroclaw
Canonical: https://ross.abutalabs.com/products/zeroclaw
Homepage: https://www.zeroclawlabs.ai/
Language: Rust
License: Apache-2.0
License Family: permissive
Topics: agent, agentic, ai, openclaw, infra, ml, os, zeroclaw
Last push: 2026-08-27T00:35:39+00:00

## Health v2 (maintenance only)
Score: 81/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 96, longevity 14
- inputs: {"age_days": 201, "days_push": 7, "days_rel": 31, "gap_med": 1, "n_releases_24m": 38}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 32660, forks 4915 (observed 2026-08-28T04:11:57.771319+00:00)

## What it is
ZeroClaw is a fast, lightweight personal AI assistant agent runtime written as a single Rust binary that you self-host. It connects to many LLM providers, communicates through 30+ channels (Discord, Telegram, Matrix, email, voice, webhooks, CLI), and acts via tools like shell, browser, HTTP, and custom MCP servers, all running locally with your own keys and data.

## Use cases
- run a self-hosted personal AI assistant on my own machine
- connect an LLM agent to Discord or Telegram
- deploy an autonomous agent that can use shell and browser tools
- swap between OpenAI, Anthropic, and Ollama providers easily
- run an AI assistant on any OS from a single binary
- keep my AI agent data and keys local instead of in the cloud

## When to choose
- you want full ownership of your agent, data, and the machine it runs on
- you need a lightweight, fast single-binary agent runtime
- you want multi-channel reach (Discord, Telegram, Matrix, email, voice, webhooks) from one agent
- you want to mix and match LLM providers and custom MCP tool servers

## When to avoid
- you want a fully managed cloud-hosted assistant with zero setup
- you need a polished GUI-first consumer assistant rather than a configurable runtime
- you prefer not to manage your own API keys and local deployment

## Facets
- artifact type: application
- maturity: active
- function: agent-framework, chatbot, cli, mcp, llm-inference, workflow-automation
- domain: large-language-models, chatbots, self-hosted, developer-tools
- platform: rust, cross-platform, windows, cli, self-hosted
- tags: personal-ai-assistant, agent-runtime, single-binary, rust, multi-channel, local-first, openclaw-fork, ai-agents, automation, linux, macos, discord, telegram

## Member repositories
- zeroclaw-labs/zeroclaw (main) score 81

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:11:57.771319+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T16:51:55.198946+00:00, confidence not recorded.
  - readme: https://github.com/zeroclaw-labs/zeroclaw (fetched 2026-08-28T04:11:57.771319+00:00, sha 81802ea7a68f)
  - homepage: https://www.zeroclawlabs.ai/ (fetched 2026-08-29T07:48:53.338216+00:00, sha 75bec110a999)
  - registry_crates: https://crates.io/api/v1/crates/zeroclaw (fetched 2026-08-29T07:48:53.360729+00:00, sha 2b5fcc69c0c8)
- Data as of 2026-08-30T08:39:29.467469+00:00.