whereismyai
MimiClaw
MimiClaw is an AI assistant that runs on an ESP32-S3 microcontroller chip using pure C to manage hardware and AI tasks without an operating system.
MimiClaw is an AI assistant compiled to bare-metal C for deployment on ESP32-S3 microcontrollers. It operates without an operating system, runtime, or server-side requirements. The system is designed for always-on performance using 0.5W of power while retaining persistent memory across reboots. Users interact with the agent via Telegram or over a local area network using WebSockets. It enables direct control of physical hardware including GPIO, sensors, and actuators. The code is open source and MIT licensed.
MimiClaw at a glance
- Best for: Software engineer
- Runs on: Web
- Pricing: Open source
- Released: February 2026
- Category: AI agent tool
- Website: mimiclaw.io
Key features
- Operates on ESP32-S3 chips
- Uses pure C without an operating system
- Maintains persistent memory in flash storage
- Supports hardware control via GPIO and sensors
- Communicates via Telegram
- Provides LAN access through WebSockets
Use it for
- Controlling physical switches and hardware components
- Monitoring environmental data like temperature
- Running AI agents on low-power, portable hardware
Alternatives to MimiClaw
- ESP-Claw — Runs agent runtimes directly on device chips for local decision making.
- Composio — Equips AI agents with tools, managed user auth, and sandboxes across 1,500+ apps
- Rakazo — Complete ownership of data, models, and execution environment.
- NumaClaw — Deploys 250+ specialized AI agents and autonomous companies on dedicated servers.
- PhoneClaw — It executes AI agent tasks and system tool calls entirely on-device.
- nanobot — Provides a compact, self-hosted agent runtime with long-horizon execution.
- Lloyal — Builds apps with embedded models that run offline without user API keys.
- BrowserOS — Runs entirely locally and lets AI agents use real user logins to complete tasks.
- Veecle — Compile and simulate firmware on actual target chips without physical hardware.
- Beeuni — Agents retain context and improve skills over time through a controlled review process.
- AgentDock — It coordinates multi-device workflows and provides secure environment access.
- Goose — Runs locally on your machine with support for 70+ extensions and any LLM.
All MimiClaw alternatives