# tnm/zclaw

Your personal AI assistant at all-in 888KiB (~35KB in app code). Running on an ESP32. GPIO, cron, custom tools, memory, and more.

Repository: https://github.com/tnm/zclaw
Canonical: https://ross.abutalabs.com/products/zclaw
Homepage: https://zclaw.dev
Language: C
License: MIT
License Family: permissive
Last push: 2026-05-17T03:43:36+00:00

## Health v2 (maintenance only)
Score: 66/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 82, release rhythm 75, longevity 14
- inputs: {"age_days": 198, "days_push": 108, "days_rel": 164, "gap_med": 4, "n_releases_24m": 4}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2221, forks 192 (observed 2026-08-28T04:06:27.613310+00:00)

## What it is
zclaw is a personal AI assistant firmware written in C that runs entirely on ESP32 microcontrollers within a strict 888 KiB all-in firmware budget. It maps natural-language chat (via Telegram or a web relay) to on-device tool calls including scheduled tasks, GPIO control, sensor reads, persistent memory, and user-defined custom tools.

## Use cases
- run an AI assistant on an ESP32 microcontroller
- control GPIO pins with natural language commands
- schedule recurring tasks like watering plants from a chat bot
- build a tiny self-contained LLM agent on embedded hardware
- remind me in 20 minutes via a Telegram bot
- read DHT11/DHT22 sensors and remember values on-device
- flash a minimal AI firmware under 1MB on ESP32-C3/S3/C6

## When to choose
- you want a self-hosted AI assistant running entirely on cheap ESP32 hardware
- you need natural-language control of GPIO, relays, or sensors without a separate server
- you care about an extremely small firmware footprint (888 KiB all-in)
- you want cron-like scheduling and persistent memory on a microcontroller

## When to avoid
- you need heavy LLM inference on-device - it relies on external OpenAI-compatible or Ollama backends
- you need a general-purpose home automation hub with broad device integrations
- your project targets non-ESP32 microcontrollers or requires large TLS feature sets
- you want a managed cloud product rather than flashing and provisioning your own board

## Facets
- artifact type: application
- maturity: active
- function: agent-framework, llm-inference, chatbot, scheduling, iot, embedded, developer-tools
- domain: artificial-intelligence, iot, embedded-systems, chatbots, hardware
- platform: embedded, iot, windows, cli
- tags: esp32, firmware, tiny-llm-agent, gpio, telegram-bot, c, freertos, esp-idf, home-automation, microcontroller, ai-agents, automation, linux, macos

## Member repositories
- tnm/zclaw (main) score 66

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:06:27.613310+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 2026-08-30T02:45:48.614721+00:00, confidence not recorded.
  - readme: https://github.com/tnm/zclaw (fetched 2026-08-28T04:06:27.613310+00:00, sha 6889cb2efff7)
  - homepage: https://zclaw.dev (fetched 2026-08-29T10:25:51.391797+00:00, sha 163752f38a0b)
  - site_page: https://zclaw.dev/getting-started.html (fetched 2026-08-29T10:25:51.401084+00:00, sha ede7ee8191cb)
  - site_page: https://zclaw.dev/changelog.html (fetched 2026-08-29T10:25:51.403214+00:00, sha c02543cbf8a5)
- Data as of 2026-08-30T08:39:29.467469+00:00.