Technical Documentation

Removie Platform & Developer Reference

The canonical handbook for integrating AI agents via Model Context Protocol (MCP), calling REST APIs, writing CLI automations, and deploying governed Remotion video pipelines.

Interactive MCP Console Removie Studio System Status

Documentation Sections

Governed Engine

Every video generated by Removie conforms to strict brand tokens and Zod schemas. LLMs cannot produce off-brand layouts.

Quickstart: MCP Server Setup

Connect Claude Desktop, Cursor, or any Model Context Protocol agent in less than 60 seconds.

Configure Claude Desktop

claude_desktop_config.json

Open your Claude Desktop configuration file (macOS: ~/Library/Application Support/Claude/claude_desktop_config.json, Windows: %APPDATA%\Claude\claude_desktop_config.json) and register the Removie server:

{
  "mcpServers": {
    "removie": {
      "command": "npx",
      "args": ["-y", "removiemcp", "mcp"],
      "env": {
        "REMOVIE_API_KEY": "dg_live_primary",
        "REMOVIE_BASE_URL": "http://localhost:3000"
      }
    }
  }
}

Configure Cursor IDE

Cursor Settings > Features > MCP

In Cursor, navigate to Cursor Settings > Features > MCP Servers and click Add New MCP Server:

Server Type

command (stdio)

Name

removie

Command

npx -y removiemcp mcp

Ask Claude or Cursor to Produce Your First Video

Once connected, simply prompt your agent naturally:

"Removie, list all available video templates and render a 9:16 vertical short about contractor marketing leaks for Ted Yeatts using brand kit 'brand_lcm'."