← Back to Skills Marketplace
Xiaozhi Claw
by
ClawMem.com
· GitHub ↗
· v1.0.0
· MIT-0
357
Downloads
1
Stars
1
Active Installs
1
Versions
Install in OpenClaw
/install xiaozhiclaw
Description
XiaoZhi AI Device (ESP32) integration for OpenClaw. Enables real-time voice interaction with your AI assistant through XiaoZhi hardware. Supports WebSocket b...
Usage Guidance
This plugin appears to implement what it claims: a local WebSocket bridge that streams Opus audio to/from a XiaoZhi ESP32 and uses Volcengine Doubao for STT/TTS. Before installing: (1) confirm you trust Volcengine (openspeech.bytedance.com) before providing DOUBAO_APP_ID/DOUBAO_ACCESS_TOKEN; treat those secrets like any API keys and keep them out of version control (.env is used). (2) Be aware the plugin opens a WebSocket port (default 8080) on the host — if you expose that port to untrusted networks it could be abused; prefer local network or firewall rules. (3) Note the minor config mismatch: plugin manifests reference other STT/TTS providers but the code uses Doubao — review and adjust config/schema if you expect different providers. (4) If you need higher assurance, review the Doubao request implementation and test the plugin in an isolated environment first.
Capability Analysis
Type: OpenClaw Skill
Name: xiaozhiclaw
Version: 1.0.0
The xiaozhiclaw skill bundle is a legitimate integration for XiaoZhi AI ESP32 hardware, enabling real-time voice interaction with OpenClaw agents. It implements a WebSocket server (src/websocket-server.ts) to handle Opus audio streaming and utilizes the Volcengine Doubao API (src/doubao-service.ts) for STT/TTS processing. The code follows standard practices, such as using environment variables for credentials and providing clear documentation of its network and audio processing requirements, with no evidence of malicious intent or suspicious behavior.
Capability Assessment
Purpose & Capability
Overall coherent: name, description, SKILL.md and code implement a WebSocket server, Opus audio handling, and Doubao STT/TTS which matches the stated purpose. Minor mismatch: openclaw.plugin.json lists sttProvider/ttsProvider enums (whisper/openai and openai/elevenlabs) but the implementation and docs require Volcengine Doubao — this is a configuration/documentation inconsistency, not an obvious security problem.
Instruction Scope
SKILL.md instructs only to run a WebSocket server, set DOUBAO_APP_ID/DOUBAO_ACCESS_TOKEN, and open port 8080. The runtime instructions and code use only WebSocket connections, local audio processing, and outbound HTTPS calls to the Doubao API. There are no instructions to read unrelated files or exfiltrate arbitrary data.
Install Mechanism
No install spec (instruction-only install). The repository includes normal npm dependencies (ws, dotenv, opusscript). No downloads from arbitrary URLs or extracts; install risk is typical for a Node plugin.
Credentials
Only Volcengine Doubao credentials are required (DOUBAO_APP_ID and DOUBAO_ACCESS_TOKEN), which is proportional to the stated need for STT/TTS. The code loads .env (dotenv) so users should be aware credentials can be stored in a local .env file. Confirm you trust the external provider (openspeech.bytedance.com) before supplying credentials.
Persistence & Privilege
always:false and the plugin starts a WebSocket server at startup via the plugin's startup hook — expected for a channel plugin. The skill does not request system/root privileges and does not modify other skills or system-wide configs.
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install xiaozhiclaw - After installation, invoke the skill by name or use
/xiaozhiclaw - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
XiaoZhiClaw v1.0.0 – Initial Release
- Adds OpenClaw channel for XiaoZhi AI ESP32 device integration.
- Supports real-time voice interaction via WebSocket (default port: 8080).
- Integrates Volcengine Doubao for speech-to-text (STT) and text-to-speech (TTS).
- Handles Opus audio streaming for efficient voice communication.
- No external API keys in code; credentials set via environment variables.
- Enables two-way AI voice interaction and basic device control through XiaoZhi hardware.
Metadata
Frequently Asked Questions
What is Xiaozhi Claw?
XiaoZhi AI Device (ESP32) integration for OpenClaw. Enables real-time voice interaction with your AI assistant through XiaoZhi hardware. Supports WebSocket b... It is an AI Agent Skill for Claude Code / OpenClaw, with 357 downloads so far.
How do I install Xiaozhi Claw?
Run "/install xiaozhiclaw" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Xiaozhi Claw free?
Yes, Xiaozhi Claw is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Xiaozhi Claw support?
Xiaozhi Claw is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Xiaozhi Claw?
It is built and maintained by ClawMem.com (@leohuang8688); the current version is v1.0.0.
More Skills