Agents, MCP and CLI

Connect Cursor, Claude, Gemini, or a terminal to Long Video Studio. Planning and scene generation still spend your credits.

1. Create a token

Sign in, open Account → Agents, and copy the viv_ token once. Do not paste it in a public chat.

Open Account

What agents can do

  • Plan a long video from a story prompt
  • See credits before you spend
  • Generate or regenerate scenes
  • Edit the timeline, then stitch for free

Replace viv_YOUR_TOKEN with the token from Account.

Cursor

~/.cursor/mcp.json or Cursor Settings → MCP

Paste, save, then reload MCP servers.

{
  "mcpServers": {
    "vivify": {
      "url": "https://api.vivifyai.app/api/studio/mcp",
      "headers": {
        "Authorization": "Bearer viv_YOUR_TOKEN"
      }
    }
  }
}

Claude Code

Terminal

Run once in the project, then restart Claude Code.

claude mcp add --transport http vivify https://api.vivifyai.app/api/studio/mcp --header "Authorization: Bearer viv_YOUR_TOKEN"

Claude Desktop

claude_desktop_config.json

Quit Claude Desktop fully, then open it again.

{
  "mcpServers": {
    "vivify": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-remote",
        "https://api.vivifyai.app/api/studio/mcp",
        "--header",
        "Authorization: Bearer viv_YOUR_TOKEN"
      ]
    }
  }
}

Gemini CLI

~/.gemini/settings.json

Save, then restart the Gemini CLI.

{
  "mcpServers": {
    "vivify": {
      "httpUrl": "https://api.vivifyai.app/api/studio/mcp",
      "headers": {
        "Authorization": "Bearer viv_YOUR_TOKEN"
      }
    }
  }
}

Windsurf

~/.codeium/windsurf/mcp_config.json

Reload MCP servers in Windsurf.

{
  "mcpServers": {
    "vivify": {
      "url": "https://api.vivifyai.app/api/studio/mcp",
      "headers": {
        "Authorization": "Bearer viv_YOUR_TOKEN"
      }
    }
  }
}

CLI / curl

Terminal

Same token. Studio plan and generate endpoints charge credits.

curl -s https://api.vivifyai.app/api/studio/catalog \
  -H "Authorization: Bearer viv_YOUR_TOKEN"

curl -s https://api.vivifyai.app/api/studio/plan \
  -H "Authorization: Bearer viv_YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"prompt":"A rainy Tokyo alley at night, 48 seconds, cinematic"}'