Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
Address security review findings: (1) default audioMode changed to 'tts' — no system audio capture by default; (2) setPermissionRequestHandler scoped to system audio mode only, approves media permissions only (not blanket true); (3) setDisplayMediaRequestHandler only registered in system audio mode. TTS mode has zero media capture footprint.
What This Skill Does
A transparent desktop overlay that renders a pixel art lobster on your screen, animating its mouth in sync with speech from a local TTS server. Runs as an Electron app bundled inside the skill. Supports Windows and Linux only.
Because the envelope data comes from the TTS server directly, mouth movement is tied only to AI speech rather than all system audio.
When to use it
- Adding a visual avatar to an AI voice agent
- Displaying an animated overlay during TTS-driven conversations
- Setting up a lip-synced desktop mascot for a home assistant
- Configuring a second-monitor overlay for a streaming AI setup
- Demoing an animated agent persona in a desktop environment
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: set up the pixel lobster on my Windows desktop
- 1Navigate to the skill's app/ directory and run npm install
- 2Open config.json and set ttsUrl to the local XTTS server address
- 3Set audioMode to tts and adjust monitor and lobsterScale as needed
- 4Launch the overlay with npx electron . or the included launch.sh script
- 5Confirm the lobster appears on screen and mouth animates during TTS playback
A transparent pixel lobster overlay running on the desktop, lip-syncing to the connected TTS server
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.