Making AI Assistants Cheap Again
中文 | English
RT-Claw is an OpenClaw-inspired intelligent assistant for embedded devices. Multi-RTOS support via OSAL. Build swarm intelligence with networked nodes. ESP32-S3 WiFi support adapted from MimiClaw.
Deploy your own AI assistant on hardware that costs just one dollar — seamlessly integrated into your daily workflow, efficiently bridging the digital and physical worlds.
Architecture · ESP32-C3 QEMU Guide · Contributing · Coding Style
rt-claw brings intelligence from the cloud to the edge through low-cost embedded nodes and swarm networking. Each node can sense the world, collaborate with others, and execute control tasks in real time.
RT-Claw exposes atomized hardware capabilities — GPIO, sensors, LCD, networking — as tools that an LLM can dynamically orchestrate. Adapt to any scenario without writing, compiling, or flashing embedded code again.
| Feature | Description | Status |
|---|---|---|
| LLM Chat Engine | Interactive conversation with Claude API over HTTP | Done |
| Tool Use | LLM-driven hardware control (GPIO, system info, LCD, audio, scheduler, HTTP) via function calling; 30+ built-in tools | Done |
| LCD Graphics | 320x240 RGB565 framebuffer with text, shapes, and drawing primitives; AI agent can draw on screen via tool calls | Done |
| OLED Display | SSD1306 I2C OLED driver for xiaozhi-xmini board | Done |
| Audio | ES8311 codec driver with preset sound effects (success, error, notify, alert); AI-controllable volume and beep | Done |
| Chat-first Shell | UART REPL with insert-mode editing, tab completion, UTF-8 support; direct input goes to AI, /commands for system | Done |
| OSAL | Write once, run on FreeRTOS and RT-Thread with zero code changes | Done |
| Gateway | Thread-safe message routing between services | Done |
| Networking | Ethernet + HTTP client on ESP32-C3 QEMU; WiFi on real hardware | Done |
| Multi-Model API | Support mainstream LLM APIs: Claude, GPT, Gemini, DeepSeek, GLM, MiniMax, Grok, Moonshot, Baichuan, Qwen, Doubao, Llama (Ollama) | Planned |
| Web Config Portal | Lightweight built-in web page for configuring API keys, selecting models, and tuning parameters at runtime | Planned |
| Swarm Intelligence | Node discovery, heartbeat, capability bitmap, remote tool invocation across nodes | In Progress |
| Conversation Memory | Short-term RAM ring buffer + long-term NVS Flash persistent storage; AI can save/delete/list memories | Done |
| Skill Memory | Nodes learn and recall frequently used operation patterns | In Progress |
| Scheduled Tasks | Timer-driven task execution and periodic automation; AI can create/list/remove tasks | Done |
| IM Integrations | Feishu (Lark) via WebSocket long connection; planned: DingTalk, QQ, Telegram | In Progress |
| Claw Skill Provider | Serve as a skill for other Claws, giving them the ability to sense and control the physical world | Planned |
+--------------------------------------------------------------+ | rt-claw Application | | gateway | swarm | net | ai_engine | tools | shell | sched | | feishu | heartbeat | lcd | audio | memory | skill | +--------------------------------------------------------------+ | osal/claw_os.h (OSAL API) | +-------------------+------------------------------------------+ | FreeRTOS (IDF) | RT-Thread | +-------------------+------------------------------------------+ | ESP32-C3 / S3 | QEMU vexpress-a9 | | WiFi / BLE / OLED | Ethernet / UART | | Audio (ES8311) | | +-------------------+------------------------------------------+
| Platform | Target | RTOS | Build | Status |
|---|---|---|---|---|
| ESP32-C3 | QEMU (Espressif fork) | ESP-IDF + FreeRTOS | Meson + CMake | AI + Ethernet verified |
| ESP32-S3 | QEMU (Espressif fork) | ESP-IDF + FreeRTOS | Meson + CMake | AI + Ethernet verified |
| ESP32-C3 | Real hardware | ESP-IDF + FreeRTOS | Meson + CMake | Untested |
| ESP32-S3 | Real hardware | ESP-IDF + FreeRTOS | Meson + CMake | Untested |
| QEMU vexpress-a9 | QEMU | RT-Thread | Meson + SCons | Boot + Ethernet verified |
Note: upstream STM32 QEMU boards were evaluated on QEMU 10.2.x but are not listed here yet because Ethernet is still unimplemented there, so rt-claw cannot provide QEMU networking on those machines today.
Requires an ESP32-S3 board with 16 MB flash and 8 MB PSRAM (e.g. ESP32-S3-DevKitC-1).
1. Install system dependencies + ESP-IDF
# Ubuntu / Debian
sudo apt install git wget flex bison gperf python3 python3-venv \
cmake ninja-build ccache libffi-dev libssl-dev dfu-util \
libusb-1.0-0 meson
# One-line ESP-IDF setup (clones ESP-IDF v5.4, installs toolchain)
./scripts/setup-esp-env.sh
2. Configure API key
source $HOME/esp/esp-idf/export.sh
idf.py -C platform/esp32s3 menuconfig
# Navigate: Component config → rt-claw Configuration → AI Engine
# → API Key / API URL / Model
3. Configure WiFi
Option A — build-time defaults (menuconfig):
idf.py -C platform/esp32s3 menuconfig
# Navigate: Component config → rt-claw Configuration → WiFi
# → Default WiFi SSID
# → Default WiFi password
Option B — runtime via shell (saved to NVS, survives reboot):
/wifi_set <SSID> <PASSWORD>
NVS credentials (Option B) take priority over build-time defaults.
If neither is configured, the device boots offline and waits for /wifi_set.
4. Build
make esp32s3
5. Flash and monitor
# Flash (auto-detects serial port)
make flash-esp32s3
# Serial monitor (Ctrl+] to exit)
make monitor-esp32s3
# Or specify a port explicitly
idf.py -C platform/esp32s3 -p /dev/ttyUSB0 flash monitor
6. Shell commands
| Command | Description |
|---|---|
| (direct input) | Send message to AI |
/ai_set key|url|model <v> | Set AI API config (persisted to NVS) |
/ai_status | Show current AI config |
/feishu_set <id> <secret> | Set Feishu credentials (reboot to apply) |
/feishu_status | Show Feishu config |
/wifi_set <SSID> <PASS> | Save WiFi credentials to NVS |
/wifi_status | Show connection state and IP |
/wifi_scan | Scan nearby access points |
/remember <key> <value> | Save fact to long-term memory |
/forget <key> | Delete fact from long-term memory |
/memories | List all long-term memories |
/task | List scheduled tasks (or /task rm <name> to remove) |
/skill | List or execute a skill |
/nodes | Show swarm node table |
/log [on|off] | Toggle log output |
/history | Show conversation message count |
/clear | Clear conversation memory |
/help | List all commands |
No hardware? No problem. Open cnb.cool/gevico.online/rtclaw/rt-claw to launch a CNB Cloud-Native IDE with all toolchains pre-installed. Build and run RT-Claw on QEMU directly in your browser.
1. Install system dependencies
# Ubuntu / Debian
sudo apt install git wget flex bison gperf python3 python3-venv \
cmake ninja-build ccache libffi-dev libssl-dev dfu-util \
libusb-1.0-0 libgcrypt20-dev libglib2.0-dev libpixman-1-dev \
libsdl2-dev libslirp-dev meson
# Arch Linux
sudo pacman -S --needed libgcrypt glib2 pixman sdl2 libslirp \
python cmake ninja gcc git wget flex bison meson
2. Install ESP-IDF + QEMU
# One-line setup (clones ESP-IDF v5.4, installs toolchain + QEMU)
./scripts/setup-esp-env.sh
3. Choose a configuration preset
| Preset | File | Shell | Feishu | Description |
|---|---|---|---|---|
| Quick Demo | sdkconfig.defaults.demo | On | Off | Interactive terminal with full AI agent |
| Feishu Bot | sdkconfig.defaults.feishu | Off | On | Headless IM bot, saves RAM |
| Default | sdkconfig.defaults | Off | Off | Minimal base for custom builds |
source $HOME/esp/esp-idf/export.sh
# Pick one:
cp platform/esp32c3/boards/qemu/sdkconfig.defaults.demo \
platform/esp32c3/boards/qemu/sdkconfig.defaults # Quick Demo
# cp platform/esp32c3/boards/qemu/sdkconfig.defaults.feishu \
# platform/esp32c3/boards/qemu/sdkconfig.defaults # Feishu Bot
idf.py -C platform/esp32c3 -B build/esp32c3-qemu/idf -DRTCLAW_BOARD=qemu set-target esp32c3
All presets include: AI engine, Tool Use, swarm heartbeat, scheduler, LCD, skills, and boot-time AI connectivity test.
4. Configure API key
Option A — environment variables (all platforms):
export RTCLAW_AI_API_KEY='<your-api-key>'
export RTCLAW_AI_API_URL='https://api.anthropic.com/v1/messages'
export RTCLAW_AI_MODEL='claude-sonnet-4-6'
Option B — ESP-IDF menuconfig:
idf.py -C platform/esp32c3 -B build/esp32c3-qemu/idf -DRTCLAW_BOARD=qemu menuconfig
# Navigate: Component config → rt-claw Configuration → AI Engine
Option C — Meson option:
meson configure build/esp32c3-qemu/meson -Dai_api_key='<your-api-key>'
5. (Optional) Configure Feishu bot
Option A — environment variables:
export RTCLAW_FEISHU_APP_ID='<your-app-id>'
export RTCLAW_FEISHU_APP_SECRET='<your-app-secret>'
Option B — ESP-IDF menuconfig:
idf.py -C platform/esp32c3 -B build/esp32c3-qemu/idf -DRTCLAW_BOARD=qemu menuconfig
# Navigate: Component config → rt-claw Configuration → Feishu (Lark) Integration
Create an app on Feishu Open Platform, enable
Event Subscription → Long Connection mode, and subscribe to
im.message.receive_v1. The device establishes a WebSocket long connection
on boot — no public IP required.
6. Build and run
# Unified build (recommended)
make build-esp32c3-qemu
# Run on QEMU
make run-esp32c3-qemu
# Or flash to real hardware (untested)
idf.py -C platform/esp32c3 -p /dev/ttyUSB0 flash monitor
# Prerequisites: arm-none-eabi-gcc, qemu-system-arm, scons, meson, ninja
# Configure via env vars (picked up at build time)
export RTCLAW_AI_API_KEY='<your-key>'
# Unified build
make vexpress-a9-qemu
# Start API proxy (RT-Thread has no TLS, proxy forwards HTTP->HTTPS)
python3 scripts/api-proxy.py https://api.anthropic.com &
# Run
make run-vexpress-a9-qemu
rt-claw/ ├── meson.build # Meson build definition (cross-compiles claw + osal) ├── meson_options.txt # Meson build options (osal backend, features, AI config) ├── Makefile # Unified build entry (make build-esp32c3-qemu / make vexpress-a9-qemu) ├── include/ # Unified public headers (aligned with claw/ and osal/) │ ├── claw/ # Public headers for claw/ │ │ ├── claw_config.h # Project configuration │ │ ├── claw_init.h # Boot entry API │ │ ├── core/ # Gateway, scheduler, service interface │ │ ├── services/ # AI, net, swarm, IM service headers │ │ ├── shell/ # Shared shell command headers │ │ └── tools/ # Tool Use framework headers │ └── osal/ # Public headers for osal/ │ ├── claw_os.h # OSAL API │ └── claw_net.h # Network abstraction ├── osal/ # OS Abstraction Layer │ ├── freertos/ # FreeRTOS implementation │ └── rtthread/ # RT-Thread implementation ├── claw/ # Platform-independent core │ ├── claw_init.c # Boot entry point │ ├── core/ # Gateway, scheduler, heartbeat │ ├── services/ai/ # LLM chat engine, memory, skill │ ├── services/net/ # Network service │ ├── services/swarm/ # Swarm intelligence │ ├── services/im/ # IM integrations (Feishu) │ ├── shell/ # UART REPL shell │ └── tools/ # Tool Use framework (GPIO, system, LCD, audio, scheduler, HTTP) ├── drivers/ # Hardware drivers (Linux-kernel style) │ ├── audio/espressif/ # ES8311 audio codec │ ├── display/espressif/ # SSD1306 OLED display │ ├── net/espressif/ # Shared WiFi driver (C3/S3) │ └── serial/espressif/ # Serial console ├── platform/ │ ├── common/espressif/ # Shared Espressif board helpers (WiFi init) │ ├── esp32c3/ # ESP32-C3 unified platform (boards/qemu/devkit/xiaozhi-xmini/) │ ├── esp32s3/ # ESP32-S3 unified platform (boards/qemu/default/) │ └── vexpress-a9/ # RT-Thread BSP (Meson + SCons) ├── vendor/ │ ├── lib/cjson/ # cJSON library │ └── os/ │ ├── freertos/ # FreeRTOS-Kernel (submodule) │ └── rt-thread/ # RT-Thread (submodule) ├── docs/ │ ├── en/ # English documentation │ └── zh/ # Chinese documentation └── scripts/ ├── api-proxy.py # HTTP→HTTPS proxy for QEMU (no TLS on RT-Thread) ├── setup-esp-env.sh # Install ESP-IDF + QEMU ├── gen-esp32c3-cross.py # Auto-generate Meson cross-file from ESP-IDF └── ...
Join the GTOC (Gevico Open-Source Community) channels:
- Discord: RT-Claw
- QQ Group: Join
- Telegram: GTOC Channel
- Bilibili: Zevorn
- WeChat: GTOC
MIT
