GuidesMeta Ads MCP

Meta Ads MCP Server

Connect Claude, ChatGPT, or any AI tool to your Meta Ads account. Manage campaigns, optimize creatives, and analyze performance with natural language.

What is the Meta Ads MCP?

The Pipeboard Meta Ads MCP (Model Context Protocol) Server is a source-available server that connects AI assistants like Claude, ChatGPT, and custom agents to the Meta Ads API. Manage campaigns, optimize ad creatives, analyze performance, and control targeting through natural language.

Instead of building API integrations from scratch, you get instant access to campaign management, analytics, targeting tools, creative upload, and more.

Key features:

30+ pre-built tools covering the full Meta Ads management lifecycle
Works with Claude Desktop, ChatGPT, Python, n8n, and any MCP client
Creative management with image upload, carousel ads, and A/B testing
Audience targeting with interests, behaviors, demographics, and geo-locations
Safety-first: campaigns created in PAUSED status by default

Getting Started

Set up in under 2 minutes

1. Sign up for a Pipeboard account:

pipeboard.co/auth/signup →

2. Connect your Meta Ads account:

pipeboard.co/connections →

3. Add the MCP server to Claude (or another MCP client):

https://mcp.pipeboard.co/meta-ads-mcp

Use this URL as a Streamable HTTP MCP server in your client. Authentication is handled automatically through your Pipeboard session.

Connect to Your AI Tool

The Meta Ads MCP works with any MCP-compatible client. Just add the server URL and start managing your campaigns with natural language.

Claude Pro / Max

Recommended
  1. 1.Open claude.ai/settings/integrations
  2. 2.Click "Add custom connector"
  3. 3.URL: https://mcp.pipeboard.co/meta-ads-mcp

ChatGPT

Developer mode
  1. 1.Enable Developer mode in ChatGPT Settings
  2. 2.Settings → Apps → Advanced → Developer mode
  3. 3.Create app with MCP URL

Claude Code

CLI / VS Code

Run this command in your terminal:

claude mcp add --transport http pipeboard-meta-ads https://mcp.pipeboard.co/meta-ads-mcp

Then type /mcp in Claude Code to authenticate with Pipeboard.

Alternative: use an API token for instant setup (no browser auth needed):

claude mcp add --transport http pipeboard-meta-ads https://mcp.pipeboard.co/meta-ads-mcp?token=pk_...

Get your token at pipeboard.co/api-tokens

Claude Desktop JSON configuration:

Add this to your Claude Desktop MCP settings (Settings → MCP Servers → Add):

claude_desktop_config.json
{
  "mcpServers": {
    "meta-ads": {
      "url": "https://mcp.pipeboard.co/meta-ads-mcp"
    }
  }
}

Tips for getting started

Browse our prompt library for ready-to-use prompts you can copy into Claude
Campaigns are created in PAUSED status by default so you can review before going live

Try these prompts

Campaign Overview

"List my Meta ad accounts and show campaign performance for the last 30 days"

Audience Research

"Search for interest targeting options related to fitness and estimate the audience size"

Creative Analysis

"Show me the creatives for my top-performing ads and suggest improvements"

Budget Optimization

"Analyze spend vs conversions across campaigns. How should I redistribute budget for better ROAS?"

Open Source

The MCP server is source-available on GitHub

Review the code, contribute features, or run your own instance. Built with Python and the official MCP SDK.

Best Practices

  • Create campaigns in PAUSED status, then review before activating
  • Batch operations when possible — use filters to reduce API calls
  • Test targeting with audience estimation before launching campaigns
  • Use breakdowns (age, gender, platform) to find top-performing segments
  • Cache insights data for reporting workflows
  • Use the Meta API rate limit of 200 calls/hour wisely — the MCP handles retries automatically

Ready to connect your AI tools to Meta Ads?

Get started with the Meta Ads MCP in under 2 minutes.