PIPEBOARD
← Back to Guides

Meta Ads MCP Server API

Complete reference for the Pipeboard Meta Ads MCP Server. Connect any AI tool to the Meta Ads API with 30+ ready-to-use functions.

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.

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

Key features:

  • 30+ pre-built tools for every Meta Ads workflow
  • Works with Claude Desktop, ChatGPT, Python, n8n, and any MCP client
  • Secure authentication with Pipeboard API tokens
  • Real-time campaign management and performance analysis
Authentication
Get your API token

1. Sign up and connect your Meta account:

pipeboard.co/auth/signup →

2. Get your API token:

pipeboard.co/api-tokens →

3. Use in your requests:

Authorization: Bearer YOUR_TOKEN
MCP Server Endpoint
https://mcp.pipeboard.co/meta-ads-mcp

All tools are accessed through this single endpoint. Configure your MCP client once and get instant access to all functions.

How to Use the MCP

The Meta Ads MCP is designed to work with MCP clients that connect large language models (LLMs) to the server. These clients enable the LLM to intelligently use the tools based on your natural language requests, handling the protocol communication automatically.

Popular MCP Clients:

  • Claude Desktop - Chat interface with native MCP support
  • ChatGPT Desktop - OpenAI's desktop app with MCP integration
  • Python (LangChain) - Build custom AI agents and automation
  • n8n - No-code workflow automation with MCP nodes

Direct Testing (Advanced)

For testing or debugging, you can call the MCP directly using the MCP protocol:

curl -X POST https://mcp.pipeboard.co/meta-ads-mcp \
  -H "Content-Type: application/json" \
  -H "Accept: application/json, text/event-stream" \
  -H "Authorization: Bearer $PIPEBOARD_API_TOKEN" \
  -d '{
    "jsonrpc": "2.0",
    "id": 1,
    "method": "tools/call",
    "params": {
      "name": "get_ad_accounts",
      "arguments": {}
    }
  }'

Replace $PIPEBOARD_API_TOKEN with your API token from pipeboard.co/api-tokens

Available Tools

30+ functions organized by category. All tools are available through the MCP server.

Account Management
Access and manage ad account information
get_ad_accounts

Get all ad accounts accessible by a user with details

get_account_info

Get detailed information about a specific ad account

get_account_pages

Get Facebook pages associated with an ad account

Campaigns
Create, read, update, and manage campaigns
get_campaigns

Get campaigns with optional filtering by status

get_campaign_details

Get detailed information about a specific campaign

create_campaign

Create a new campaign with budget, objective, and targeting

update_campaign

Update campaign settings and budgets

create_budget_schedule

Schedule budget increases for high-demand periods

Ad Sets
Manage ad sets with targeting and budgets
get_adsets

Get ad sets with optional campaign filtering

get_adset_details

Get detailed information about an ad set

create_adset

Create a new ad set with targeting and optimization settings

update_adset

Update ad set settings including frequency caps and budgets

Ads
Create and manage individual ads
get_ads

Get ads with optional campaign or ad set filtering

get_ad_details

Get detailed information about a specific ad

create_ad

Create a new ad with an existing creative

update_ad

Update ad status, bid, or creative

Creatives & Media
Upload images and manage ad creatives
get_ad_creatives

Get creative details for a specific ad

upload_ad_image

Upload an image to use in creatives (returns image hash)

create_ad_creative

Create a new creative with image, copy, and CTA

update_ad_creative

Update creative copy, headlines, or CTA

get_ad_image

Download and visualize ad images

Analytics & Insights
Access performance data and metrics
get_insights

Get performance metrics for campaigns, ad sets, or ads

Supports time ranges (today, last_7d, last_30d, etc.) and breakdowns (age, gender, platform, etc.)

Audience Targeting
Research and validate targeting options
search_interests

Search for interest targeting options

get_interest_suggestions

Get related interest suggestions for existing interests

estimate_audience_size

Estimate reach for targeting specifications

search_behaviors

Search for behavior targeting options

search_demographics

Search demographic targeting options (life events, income, etc.)

search_geo_locations

Search for geographic targeting (countries, regions, cities)

Premium Tools (Paid Plans Only)
Advanced duplication features for A/B testing and scaling
duplicate_campaign

Duplicate entire campaigns with all ad sets and ads

duplicate_adset

Duplicate ad sets with optional targeting changes

duplicate_ad

Duplicate ads to different ad sets

duplicate_creative

Duplicate creatives with copy variations

Usage Examples
Connect the MCP server to your favorite tools

Python + LangChain

Build custom automation scripts with AI agents

View Python guide →

Claude Desktop

Use natural language to manage campaigns

Setup in Claude →

n8n Automation

Build no-code workflows for ad operations

View n8n guide →

ChatGPT

Enable MCP support in ChatGPT Desktop

Setup in ChatGPT →
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.

Rate Limits & Best Practices

Meta API Limits

The Meta Ads API has rate limits of 200 calls per hour per user. The MCP server automatically handles retries with exponential backoff.

Best Practices

  • Batch operations when possible (use filters to reduce API calls)
  • Cache insights data for reporting workflows
  • Use paused status when creating campaigns to review before activation
  • Test targeting with audience estimation before launching campaigns
Support & Documentation

Need help?