Free SKILL.md scraped from GitHub. Clone the repo or copy the file directly into your Claude Code skills directory.
npx versuz@latest install vivekkarmarkar-claude-code-os-skills-robo-voice-launch-protocolgit clone https://github.com/VivekKarmarkar/claude-code-os.gitcp claude-code-os/SKILL.MD ~/.claude/skills/vivekkarmarkar-claude-code-os-skills-robo-voice-launch-protocol/SKILL.md# Robo Voice Launch Protocol
Launch the Robo Voice infrastructure and print the connection protocol.
Use when the user says "launch robo", "start robo voice", "robo voice protocol", or wants to connect Robo Voice to a Claude Code session.
## What it does
### Step 1: Launch the agent and frontend
Start both servers in the background from the livekit-project directory:
```bash
cd ~/Python\ Files/livekit-project/python-agents-examples/complex-agents/avatars/anam/agent-py
nohup uv run python src/agent.py dev > /tmp/robo-agent.log 2>&1 &
```
```bash
cd ~/Python\ Files/livekit-project/python-agents-examples/complex-agents/avatars/anam/frontend
nohup pnpm dev > /tmp/robo-frontend.log 2>&1 &
```
Wait a few seconds, then verify both are running by checking the logs. Extract the frontend URL from the log (look for "Local:" line).
### Step 2: Print the protocol
After confirming both servers are running, print the following to screen:
```
═══════════════════════════════════════════════════════
ROBO VOICE — LAUNCH PROTOCOL
═══════════════════════════════════════════════════════
Agent: ✓ Running
Frontend: ✓ Running → <frontend URL>
NEXT STEPS:
1. Exit this Claude Code session
2. cd into your target project folder
3. Launch Claude Code with channels:
claude --dangerously-load-development-channels server:robo-voice
4. Open <frontend URL> → select claude-code → connect → talk
SESSION FLAGS (re-pass channels flag every time):
--name <name> Name a session
--continue Continue most recent session
--resume Pick from session list
═══════════════════════════════════════════════════════
```
Replace `<frontend URL>` with the actual URL extracted from the frontend log.
## Rules
- Kill any existing agent/frontend processes before starting new ones
- Kill anything on port 7890 before starting (old MCP servers)
- Always verify both servers started successfully before printing the protocol
- If either server fails to start, report the error instead of printing the protocol