clawtime

What This Skill Does

ClawTime is a sophisticated webchat interface designed specifically for OpenClaw agents. It elevates the standard text-based interaction by integrating 3D avatars, secure passkey authentication, and high-fidelity voice mode. This skill transforms your OpenClaw experience into a personal interactive companion, providing a visual representation of your AI and allowing for natural, spoken communication. By leveraging local whisper.cpp speech-to-text, ClawTime ensures that your interactions remain private and efficient while maintaining high accuracy in transcription.

Installation

Installation is a multi-step process. First, navigate to your workspace directory (~/.openclaw/workspace) and clone the repository using git clone https://github.com/youngkent/clawtime.git. Install dependencies via npm install.

For the optional but recommended Voice Mode, you must install whisper.cpp. Clone the repository, build it using make, and download the base.en or base model. Create a wrapper script at /usr/local/bin/whisper-transcribe to handle the file conversion and transcription. Finally, configure your ~/.clawtime/.env file by inserting your GATEWAY_TOKEN, BOT_NAME, and BOT_EMOJI to ensure the interface can authenticate with the OpenClaw gateway correctly.

Use Cases

ClawTime is ideal for users who want a more human-like connection with their AI assistant. It is perfect for accessibility, allowing users to communicate via voice instead of typing. Developers and enthusiasts can use it to build interactive kiosks, virtual receptionists, or personalized assistants that match specific personality profiles. The customizable 3D avatar capability makes it a powerful tool for branding your OpenClaw agent for professional or creative projects.

Example Prompts

1. "OpenClaw, launch ClawTime and set my assistant's avatar to a friendly blue robot named Nexus."
2. "I need to update my agent's personality; please configure ClawTime to reflect a formal and analytical tone with the 🦉 emoji."
3. "Hey, help me verify if my Whisper STT is configured correctly for voice mode, and if not, walk me through the setup for the base model."

Tips & Limitations

Always ensure your GATEWAY_TOKEN is current; the interface will fail to launch without valid identification. While the 3D avatar feature is visually immersive, it requires a modern web browser to render smoothly. If you experience latency in voice mode, consider using a smaller whisper model. Note that the browser-based speech recognition API is a fallback only and will not perform as accurately as the server-side Whisper implementation. Always test your audio input settings in your OS before attempting to use the voice mode features.