irinabuht12-oss/google-ads-meta-ads-mcp

A Google Ads MCP server, a Meta Ads MCP server (Facebook and Instagram ads), Google Analytics 4 and Google Search Console in one remote MCP for Claude, Claude Code, Claude Desktop, ChatGPT, Cursor, Windsurf and n8n: 250+ tools, OAuth login, no API keys or developer token, approval-gated writes, 7-day free trial.

Skip the setup: use the hosted Ryze MCP (no developer token, no Google Cloud project, approval-gated writes).

Connect Ryze in one click →

• Google Ads MCP: tools, prompts, install per client, comparison (product page)
• Google Analytics MCP server: GA4 reports, realtime, AI-referral traffic from ChatGPT, Claude, Gemini and Perplexity (product page)
• Meta Ads MCP: Facebook and Instagram Ads tools, prompts, install per client (product page)
• Agent-readable index: llms.txt

claude mcp add ryze --transport http https://connector.get-ryze.ai/mcp

claude.ai, Claude Desktop, ChatGPT, Cursor: Settings › Connectors › Add custom connector, paste https://connector.get-ryze.ai/mcp, sign in with the Google or Facebook account that owns the ads. No API keys, no developer token, nothing to install. Screenshots for every client →

See how to run a full Google Ads and Meta Ads account audit using Claude with MCP — from data pull to automated audit report and presentation, in minutes instead of hours.

What you'll see: Connect your ad accounts → Claude analyzes campaign performance, budget efficiency, and targeting gaps across Google Ads & Meta Ads → generates a complete audit report with recommendations → builds a ready-to-share presentation. Try it free on Ryze AI →

One MCP server. Three platforms. 250+ tools. Manage all your advertising and analytics from ChatGPT, Claude, Cursor, n8n, Windsurf, and more. Powered by Ryze AI.

The Model Context Protocol (MCP) is the open standard that lets AI assistants like Claude and ChatGPT securely use external tools. This repo is a Google Ads MCP, a Meta Ads MCP, and a GA4 MCP in one server: it exposes the Google Ads API, Meta Marketing API, and Google Analytics 4 Data API as MCP tools, so you create campaigns, research keywords, upload creatives, build audiences, and pull performance reports by chatting with your AI assistant instead of clicking through three dashboards.

You can run this server next to the official ones; they do not conflict. Use the official Google server when you need a local, read-only, code-first setup, and this one when you want Google Ads, Meta Ads and analytics together with writes.

Is there an official Google Ads MCP server? Yes. Google publishes googleads/google-ads-mcp: open source, Python, read-only, and it needs a developer token and a Google Cloud project. This repo is the hosted alternative with writes, Meta Ads and GA4 included.

Is there a free Google Ads MCP? Google's official server (googleads/google-ads-mcp) is free and open source, but read-only and it needs a developer token and a Google Cloud project. This connector is included in every Ryze plan; Paid Ads is $89/month with a 7-day free trial, so you can try it at no cost.

Does it work as a Google Ads MCP for Claude Desktop, Claude Code, Cursor and ChatGPT? Yes, one URL for all of them. Claude Code: claude mcp add ryze --transport http https://connector.get-ryze.ai/mcp. claude.ai, Claude Desktop, ChatGPT and Cursor: Settings › Connectors › Add custom connector › paste the URL.

How do I connect Claude to Google Ads or Meta Ads? Add the connector URL, sign in with the Google or Facebook account that owns the ad accounts, and ask Claude for a campaign summary. Screenshots for every client: setup guide.

Do I need a Google Ads developer token or a Meta app? No. Login is OAuth through Ryze; there is nothing to install and no API key to manage.

Can Claude manage or optimize my Google Ads and Meta ads through this? Yes. Reads are immediate; writes (budgets, bids, pausing, new campaigns, keyword changes) are staged and applied only after you approve them.

What is the difference between a Facebook Ads MCP and a Meta Ads MCP? The same thing: Facebook and Instagram ads run on Meta's Marketing API, which is what the Meta Ads tools here expose.

Is it a remote MCP or a local one? Remote, over Streamable HTTP. Nothing runs on your machine.

Does it work for agencies with many Google Ads or Meta accounts? Yes. listAccessibleCustomers returns every Google Ads account the login can reach and you pass customerId per call (manager accounts pass loginCustomerId); Meta works the same way through listAdAccounts.

Which tools are included? 150+ Google Ads tools, 80+ Meta Ads tools, 20+ GA4 tools. The full list is in All Available Tools; per-platform pages: Google Ads MCP and Meta Ads MCP.

Marketers and agencies juggle Google Ads, Meta Ads, and GA4 daily — each with its own dashboard, API, and workflow. Existing MCP servers only cover one platform at a time, forcing you to install and configure multiple tools.

This server unifies everything:

• Campaign Management — Create, update, pause, and delete Search, Display, Shopping, PMax, Demand Gen, Video, and App campaigns
• Keyword Research — Keyword Planner integration with volume, CPC, and competition data
• Bidding & Budgets — Portfolio strategies, shared budgets, bid adjustments by device/location/schedule
• Ad Creation — Responsive Search Ads, Display Ads, Demand Gen Ads, App Ads with full creative control
• Audience Targeting — Custom audiences, remarketing lists, customer match, combined audiences
• Performance Analytics — Campaign/ad group/keyword/ad-level metrics, search terms, ROAS calculations
• Extensions — Sitelinks, callouts, calls, images, prices, promotions, structured snippets, lead forms
• Experiments — A/B testing with experiment creation, scheduling, and promotion
• Negative Keywords — Campaign-level, shared sets, auto-suggestions
• Labels & Organization — Create and apply labels across campaigns, ad groups, and ads
• Conversion Tracking — Create actions, upload conversions, custom variables, value rules

• Campaign CRUD — Create, read, update, and delete campaigns with full objective support
• Ad Set Management — Budget, schedule, placement, and optimization controls
• Creative Management — Ad creatives, image/video upload, carousel and collection formats
• Audience Targeting — Custom audiences, lookalikes, interest/behavior/demographic targeting
• Lead Generation — Lead forms, lead retrieval, test leads
• Product Catalogs — Catalog management, product feeds, product sets
• Performance Insights — Account/campaign/ad set/ad-level metrics with breakdowns
• Page Management — Page posts, promotable posts, page insights, boosted posts
• Business Management — Business info, users, pages, Instagram accounts, pixels

• Reporting — Run standard reports, pivot reports, and realtime reports
• Audience Analysis — List audiences, channel groups, and segments
• Property Configuration — Data streams, enhanced measurement, data retention settings
• Attribution — Attribution settings and model configuration
• Key Events — List and analyze conversion events
• Integrations — BigQuery links, DV360 links, Search Ads links, Firebase links
• Annotations — Property-level annotations for tracking changes
• Metadata — Available dimensions, metrics, and compatibility checks

• Compare Google Ads vs Meta Ads campaign performance side by side
• Correlate ad spend with GA4 conversion data
• Unified ROAS analysis across platforms
• Cross-platform budget allocation recommendations

Remote OAuth servers are added as connectors, not in claude_desktop_config.json:

1. Settings › Connectors › Add custom connector
2. Name Ryze, URL https://connector.get-ryze.ai/mcp, click Add
3. Sign in with Google (Google Ads, GA4, Search Console) or Facebook (Meta Ads) and pick the accounts

claude mcp add ryze --transport http https://connector.get-ryze.ai/mcp

See the full Claude Setup Guide for detailed instructions.

1. Go to Settings > Connectors > Add custom connector
2. Enter the MCP endpoint URL: https://connector.get-ryze.ai/mcp
3. Name it Google Meta Ads GA4
4. Click Save

See the full ChatGPT Setup Guide.

Add to your MCP config (~/.cursor/mcp.json or ~/.codeium/windsurf/mcp_config.json):

{
  "mcpServers": {
    "google-meta-ads-ga4": {
      "url": "https://connector.get-ryze.ai/mcp"
    }
  }
}

Use the MCP Server Trigger node with Streamable HTTP:

1. Add an MCP Client node to your workflow
2. Set the Server URL to: https://connector.get-ryze.ai/mcp
3. Enable Bearer authentication (recommended)

See the full n8n Setup Guide or import the ready-made workflow template.

claude mcp add google-meta-ads-ga4 --transport http https://connector.get-ryze.ai/mcp

"Show me all my Google Ads campaigns and their performance this month"
"Create a new Search campaign targeting 'best CRM software' keywords with a $50/day budget"
"What are my top 10 keywords by conversions in the last 30 days?"
"Pause all campaigns with ROAS below 2.0"
"Add negative keywords 'free' and 'cheap' to all my Search campaigns"
"Generate keyword ideas for 'project management tools' with volume data"

"List all active Meta ad campaigns and their spend this week"
"Create a lookalike audience based on my top purchasers"
"What's the CPL for my lead generation campaigns on Facebook?"
"Upload this image and create a new ad creative for my awareness campaign"
"Show me ad performance broken down by age and gender"
"Boost my latest page post with a $100 budget targeting US users"

"Show me a realtime report of active users on my website right now"
"What are my top traffic sources by conversions this month?"
"List all key events configured in my GA4 property"
"Run a report comparing organic vs paid traffic for the last 90 days"
"What custom dimensions and metrics do I have set up?"

"Compare my Google Ads and Meta Ads ROAS for Q1 2026"
"Which platform is driving more conversions per dollar spent?"
"Show me all campaigns across Google and Meta with spend over $1000 this month"
"Correlate my GA4 conversion events with Google Ads campaign performance"

+------------------+     +------------------------+     +------------------+
|                  |     |                        |     |                  |
|  ChatGPT         |     |                        |     |  Google Ads API  |
|  Claude          |     |   MCP Server           |     |                  |
|  Cursor          +---->+   (Remote/Hosted)       +---->+  Meta Ads API    |
|  n8n             |     |                        |     |                  |
|  Windsurf        |     |   250+ Tools           |     |  GA4 Data API    |
|  Codex           |     |   OAuth 2.1 + PKCE     |     |                  |
|                  |     |                        |     |                  |
+------------------+     +------------------------+     +------------------+
    AI Assistants           Model Context Protocol         Ad & Analytics
                                                            Platforms

Your AI assistant communicates with this MCP server using the Model Context Protocol. The server handles authentication and routes requests to the appropriate Google Ads, Meta Ads, or GA4 API.

• OAuth 2.1 with PKCE — Industry-standard authentication
• No data storage — The server does not store your ad data or credentials
• Your tokens stay local — Authentication tokens are stored on your machine only
• Read-only by default — Write operations require explicit confirmation
• Scoped access — Request only the permissions you need

Contributions are welcome! Please see our issue templates to report bugs or request features.

1. Fork this repository
2. Create your feature branch (git checkout -b feature/amazing-feature)
3. Commit your changes (git commit -m 'Add amazing feature')
4. Push to the branch (git push origin feature/amazing-feature)
5. Open a Pull Request

If this project helps you manage ads more efficiently, please give it a star! It helps others discover it.

This project is licensed under the MIT License — see the LICENSE file for details.

Google Ads MCP · Meta Ads MCP · Facebook Ads MCP · GA4 MCP · Model Context Protocol
Built for marketers, agencies, and developers who want AI-powered ad management.

🚀 Get started with Ryze AI