mimoLive API Reference

A comprehensive reference for the mimoLive HTTP API and WebSocket protocol, covering every endpoint, response format, and common recipe — verified against a live mimoLive instance.

This reference is designed to work with AI coding assistants like Claude, Cursor, and GitHub Copilot. Drop it into your project context and your AI assistant will know exactly how to write correct mimoLive API calls.

What's Covered

• Documents — list, modify, start/stop shows, get program output, control the built-in record/stream/playout/fullscreen outputs
• Layers & Variants — create/delete layers, toggle live, modify parameters, cycle variants, trigger signals
• Sources — create/delete, modify, preview images, media control, signals, and source-type actions
• Output Destinations — create/delete, file recording, RTMP streaming, NDI, fullscreen, and more
• Layer Sets — create, update, delete, and recall preset layer configurations
• Data Stores — store, read, and delete raw per-document blobs
• Types — discover layer / source / output-destination type identifiers and their parameters
• Devices — list available video/audio devices
• Comments — inject comments into the live comment system
• Accounts — list configured web-service accounts (Zoom, YouTube, …)
• Zoom Meetings — join/leave (account required), participants, source assignment, 20+ meeting actions
• WebSocket — real-time state change events with connection examples
• MCP — native Model Context Protocol endpoint for AI agents
• mlController Proxy — simplified API on port 8990 with optional auth

Quick Start

As a Claude Code plugin (recommended)

Install the reference as a plugin from its own public marketplace. The mimolive-http-api skill then loads automatically whenever you work on a mimoLive integration:

/plugin marketplace add boinx/mimoLive-API-Reference
/plugin install mimolive-api@mimolive

The plugin version tracks the mimoLive release line (e.g. 6.19.0); run /plugin marketplace update mimolive to pull the latest. This marketplace is public and independent of any other marketplaces you have installed.

With Claude Code (manual)

Copy the reference into your project:

# Option A: Add to your .claude directory
mkdir -p .claude
cp mimoLive-API.md .claude/

# Option B: Reference in your CLAUDE.md
echo "For mimoLive API details, see mimoLive-API.md" >> CLAUDE.md

Then ask Claude to help you build:

Help me write a script that toggles the lower third layer when I press a hotkey

With other AI assistants

Paste the contents of mimoLive-API.md into your conversation or project context.

With curl

The reference includes working curl examples. Here's a taste:

# List open documents
curl http://localhost:8989/api/v1/documents

# Start the show
curl http://localhost:8989/api/v1/documents/{DocID}/setLive

# Toggle a layer
curl http://localhost:8989/api/v1/documents/{DocID}/layers/{LayerID}/toggleLive

# Create a live-streaming output destination
curl -X POST -H "Content-Type: application/vnd.api+json" \
  -d '{"data": {"attributes": {"output-destination-type": "com.boinx.mimoLive.outputDestination.liveStreaming", "settings": {"rtmpurl": "rtmp://...", "streamingkey": "your-key"}}}}' \
  http://localhost:8989/api/v1/documents/{DocID}/output-destinations

# Join a Zoom meeting
curl "http://localhost:8989/api/v1/zoom/join?meetingid=123456789&displayname=mimoLive"

Using with mlController

mlController is an open-source macOS menu bar app that provides a simplified proxy API for mimoLive on port 8990. The reference documents both the native mimoLive API and the mlController proxy endpoints.

If you're using Claude Code with the mlController repo, the /mimolive slash command loads this reference automatically.

Prerequisites

• mimoLive installed and running on macOS
• HTTP Server enabled in mimoLive Preferences > Remote Control
• API accessible at http://localhost:8989/api/v1

Official Documentation

• mimoLive HTTP API
• mimoLive Endpoints
• mimoLive Data Types

Community

Join the mimoLive Experts community on Discord:

Join the Discord

License

This project is licensed under the MIT License — see the LICENSE file for details.