NullClaw
Fastest, smallest, and fully autonomous AI assistant infrastructure written in Zig observed · 2026-08-28
Health v2 · maintenance only
75/100
- Activity 93
- Release rhythm 86
- Longevity 14
How is this computed?
round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10) — computed 2026-09-03. Adoption (stars, forks) is never an input.
- gap_med: 1.0
- age_days: 198
- days_rel: 96
- days_push: 45
- n_releases_24m: 29
Adoption not part of the score
8052 stars · 943 forks observed · 2026-08-28
What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded
NullClaw is an autonomous AI assistant/agent runtime written in Zig, distributed as a tiny static binary (~678 KB, ~1 MB RAM, <2 ms startup) with no external dependencies. It bundles 50+ LLM providers, 19 messaging channels, 35+ tools, pluggable memory engines, sandboxing, MCP support, and a CLI/gateway operator surface.
Use cases
- run an autonomous AI agent on a cheap ARM board or microcontroller
- self-host a personal AI assistant that works over Telegram, Discord, or Slack
- let an agent refactor code, run tests, and open PRs from the terminal
- route one assistant across OpenAI, Anthropic, Ollama, and local models
- give a chatbot persistent memory across sessions
- run an AI agent with strict sandboxing and allowlists on an edge device
When to choose
- you need an extremely lightweight, zero-dependency agent runtime for edge or low-resource hardware
- you want a single self-hosted binary that multiplexes many LLM providers and chat channels
- security matters: you need sandboxing, pairing, allowlists, and encrypted secrets by default
- you prefer fast startup and minimal memory footprint over a heavyweight framework
When to avoid
- you need a rich GUI-first experience out of the box (the NullHub UI layer is still in beta)
- you want to extend the runtime in Python or JavaScript rather than Zig
- you need mature integrations or a large plugin ecosystem beyond the built-in tools and channels
- your team requires long-term enterprise support or a large community around the tool
Facets
application · maturity active
agent-framework llm-inference chatbot cli mcp rag scheduling webhook security http-server large-language-models chatbots developer-tools self-hosted iot windows cli cross-platform self-hosted embedded iot zig single-binary autonomous-agents multi-provider memory-engines messaging-channels sandboxing edge-computing tiny-binary subagents ai-agents command-line automation linux macos docker
7 sources
- readme: https://github.com/nullclaw/nullclaw · fetched 2026-08-28 · 3b6dae480c25
- homepage: https://nullclaw.io · fetched 2026-08-29 · 073515b7c2e6
- site_page: https://nullclaw.io/nullclaw/docs/getting-started · fetched 2026-08-29 · e53531489bd4
- site_page: https://nullclaw.io/nullclaw/docs/providers · fetched 2026-08-29 · dc7a6738ab4b
- site_page: https://nullclaw.io/nullclaw/docs/tools · fetched 2026-08-29 · dc7d0dbba6be
- site_page: https://nullclaw.io/nullclaw/docs/channels · fetched 2026-08-29 · 5a1079119459
- site_page: https://nullclaw.io/nullclaw/docs/memory · fetched 2026-08-29 · 67f858ee2eef
Member repositories
| Repository | Role | Health v2 |
|---|---|---|
| nullclaw/nullclaw | main | 75 |
| nullclaw/nullhub | plugin | 75 |
For agents
markdown · JSON · MCP: product_card(name="nullclaw/nullclaw")
Data as of 2026-08-30T08:39:29.467469+00:00 · Report a problem