# nanocoai/nanoclaw

A lightweight alternative to OpenClaw that runs in containers for security. Connects to WhatsApp, Telegram, Slack, Discord, Gmail and other messaging apps,, has memory, scheduled jobs, and runs directly on Anthropic's Agents SDK

Repository: https://github.com/nanocoai/nanoclaw
Canonical: https://ross.abutalabs.com/products/nanoclaw
Homepage: https://nanoclaw.dev
Language: TypeScript
License: MIT
License Family: permissive
Topics: ai-agents, ai-assistant, claude-code, claude-skills, openclaw
Last push: 2026-08-26T21:09:22+00:00

## Health v2 (maintenance only)
Score: 82/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 99, longevity 15
- inputs: {"age_days": 214, "days_push": 7, "days_rel": 9, "gap_med": 10.5, "n_releases_24m": 7}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 30626, forks 12839 (observed 2026-08-28T04:11:54.544998+00:00)

## What it is
NanoClaw is a lightweight, self-hosted personal AI assistant that runs each agent inside its own isolated Docker container for OS-level security. It connects to messaging apps like WhatsApp, Telegram, Slack, Discord, and iMessage, and supports memory, scheduled jobs, and multiple LLM providers on a small, auditable codebase.

## Use cases
- run a personal AI assistant I can text from WhatsApp or Telegram
- self-host an AI agent with container sandboxing instead of app-level permission checks
- give each team member their own Slack bot agent with isolated memory
- schedule recurring AI tasks like morning briefings or repo issue summaries
- run a lightweight alternative to OpenClaw I can fully understand and customize
- connect one AI agent to multiple chat channels with shared or separate memory

## When to choose
- you want a personal AI agent reachable from messaging apps you already use
- security through OS-level container isolation matters to you
- you value a small, auditable codebase over a feature-heavy monolith
- you want per-agent workspaces, memory, and scheduled jobs out of the box
- you prefer self-hosting with Claude, Codex, OpenCode, or Ollama as the model

## When to avoid
- you need a managed cloud service with no self-hosting
- you require a web dashboard or GUI for administration
- you need a massive ecosystem of prebuilt integrations rather than adding skills yourself
- you cannot run Docker or WSL2 on your machine

## Facets
- artifact type: application
- maturity: active
- function: agent-framework, chatbot, scheduling, security, self-hosted, chat-interface, llm-inference
- domain: chatbots, large-language-models, self-hosted, messaging-platforms, developer-tools
- platform: windows, self-hosted
- tags: personal-ai-assistant, container-isolation, openclaw-alternative, anthropic-agents-sdk, multi-channel-messaging, claude-code, sandboxed-agents, whatsapp, telegram, slack, discord, ai-agents, automation, docker, linux, macos, typescript, nodejs

## Member repositories
- nanocoai/nanoclaw (main) score 82

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:11:54.544998+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 2026-08-29T16:52:44.698223+00:00, confidence not recorded.
  - readme: https://github.com/nanocoai/nanoclaw (fetched 2026-08-28T04:11:54.544998+00:00, sha 58c6975e0efb)
  - homepage: https://nanoclaw.dev (fetched 2026-08-29T07:49:52.619088+00:00, sha aa5ffa74c2dc)
  - site_page: https://docs.nanoclaw.dev (fetched 2026-08-29T07:49:52.639556+00:00, sha d8206ce746ef)
- Data as of 2026-08-30T08:39:29.467469+00:00.