MCP · works with Claude

Drive the studio from inside Claude

Connect Vidsly and ask Claude in plain words — a narrated voiceover MP3 or a lip-synced presenter video comes back in the chat. Same voices and render pipeline as the studio, no browser tab required.

1Get your API key

Create a key on your dashboard (starts with ksk_live_) and pick a plan — each presenter video costs 500 credits.

2Add to claude_desktop_config.json

Claude Desktop → Settings → Developer → Edit Config, then paste (with your key):

{
  "mcpServers": {
    "vidsly": {
      "command": "npx",
      "args": ["-y", "vidsly-mcp"],
      "env": { "VIDSLY_API_KEY": "ksk_live_..." }
    }
  }
}

3Just ask Claude

"Narrate this script in Aria's voice, then send me the MP3."

The file or watch link comes back in the chat. Voices, presenters and reference photos all work exactly like they do in the studio.

Presenter videos — 500 credits each from your plan's balance, refunded automatically if a render fails.
Voiceovers too — narrate a script to MP3 with 50+ voices across 15 languages (1 credit per 10 characters).
Under the hood vidsly-mcp on npm, plus a plain REST API.

Using claude.ai in the browser? A no-install connector (sign in, no key) is coming soon.