Overview

Connect your AI tools to Adsumo via the Model Context Protocol (MCP) and generate ads programmatically.

What is MCP?

The Model Context Protocol (MCP) is an open standard that lets AI tools -- like Claude Code, Cursor, and VS Code Copilot -- connect to external services and use their capabilities directly. Instead of copy-pasting between tools, your AI assistant can call Adsumo's ad generation features on your behalf.

How Adsumo MCP Works

Adsumo exposes an MCP server at:

text
https://www.adsumo.ai/api/mcp

Once connected, your AI tool gets access to 22 tools for generating video ads, image ads, voiceovers, and more -- the same capabilities available in the Adsumo dashboard and REST API.

Your AI assistant can:

  • Generate AI videos with Sora 2, Veo 3.1, or Kling 2.6
  • Create talking avatar videos with lip-sync
  • Generate product images and styled product photography
  • Create voiceovers with ElevenLabs voices
  • Browse the Meta Ad Library for competitor research
  • Manage products, avatars, and check credit balances

Authentication

MCP uses the same API keys as the REST API. Create a key in the Adsumo dashboard under Settings > API Keys, then include it as a Bearer token in the MCP connection config.

API keys are prefixed with adsumo_sk_ and are tied to a specific team. All generations consume credits from that team's balance.

API access (including MCP) requires a Starter subscription tier or above.

Client Setup

Any MCP-compatible client can connect to Adsumo -- the server uses standard streamable HTTP transport. Point your client to https://www.adsumo.ai/api/mcp with a Bearer token in the Authorization header.

We provide ready-made config snippets for these tools in the Connection guide:

ClientNotes
Claude CodeNative HTTP support
Claude DesktopRequires npx mcp-remote bridge
CursorNative HTTP support
VS Code CopilotNative HTTP support
CodexConfig via TOML
WindsurfNative HTTP support

Quick Start

1. Create an API key

Go to Settings > API Keys in the Adsumo dashboard and create a new key.

2. Add the MCP config to your AI tool

See the Connection guide for exact config snippets for each client.

3. Start using tools

Ask your AI assistant to generate ads. For example:

"Generate a 16:9 AI video with Sora 2 showing a product showcase with dynamic lighting"

"Create a voiceover with a professional female voice saying 'Try our new product today'"

"Generate 4 product images of a coffee mug on a minimalist desk"

Credits & Rate Limits

MCP uses the same credit system and rate limits as the REST API. Each generation type has a specific credit cost, and your team's balance is shared across the dashboard, API, and MCP.

Check your balance at any time by asking your AI tool to use the get_credits tool, or see the API docs for credit cost details.

Rate limits:

  • Read operations: 60 requests/minute
  • Video generation: 10 requests/minute
  • Image generation: 20 requests/minute
  • Voiceover generation: 20 requests/minute
  • Ad Library scraping: 5 requests/minute