# How to Automate Meta Ads with Claude AI and MCP — Real Workflows, Real Results

> Meta shipped an official MCP server at mcp.facebook.com/ads in April 2026 — no Developer App, no App Review, just Business OAuth. Here is how to connect Claude to your Facebook and Instagram campaigns, which of the seven OAuth scopes actually matter, and the workflows that pay for themselves.

**Source:** https://mcpplaygroundonline.com/blog/automate-meta-ads-claude-ai-mcp  
**Author:** Nikhil Tiwari  
**Published:** 2026-02-27  
**Updated:** 2026-08-17  
**Category:** Automation  
**Reading time:** 15 min read

---

📣 Meta Ads Agent — try it in your browser

Audit creatives, compare audiences, and surface budget leaks through chat — powered by **Pipeboard** (third-party Meta Business Partner, free to start).

[Try the Agent →](/templates/meta-ads-pipeboard)

📖 TL;DR — Key Takeaways

-   The **Meta Ads MCP server** gives Claude live, authenticated access to your Facebook and Instagram ad accounts
-   25 real tools — campaign creation, creative management, audience discovery, live performance insights
-   Advolve reported **90% reduction in operational work** and **15% ROAS increase** (verified Anthropic case study)
-   Setup takes 5 minutes with Pipeboard (managed) or 30–60 minutes self-hosted (free, open source)
-   Works with Claude Desktop — no custom code required to get started

🔄 Updated August 2026 — Meta shipped an official MCP server

-   **The big one: Meta now runs its own MCP server.** On 29 April 2026 Meta launched a first-party hosted endpoint at `https://mcp.facebook.com/ads`. This guide has been restructured around it — the community servers are still here, but they are no longer the default answer.
-   **No Developer App, no App Review.** The official server authenticates with standard Business OAuth. The four-to-six week App Review wait that used to be the single biggest reason marketers gave up on this is gone.
-   **Not just Meta.** Google Ads (28 April), TikTok (13 May) and X (30 June) all shipped official MCP servers in the same window. The third-party era for ad-platform MCP is closing fast — see [our X guide](/blog/x-twitter-ads-mcp-claude-automation-guide) for that one.
-   **Test any of it free** in [MCP Agent Studio](/mcp-agent-studio) — paste a server URL, pick from 60+ models, and watch every tool call live. No install.

⚠️ Know which server you are connecting to

This guide covers **two different kinds of server**. The endpoint at `mcp.facebook.com/ads` is **operated by Meta Platforms, Inc.** and authenticates through Meta's own Business OAuth. Everything else here — including **pipeboard-co/meta-ads-mcp** — is built by **independent developers and open-source contributors**, is not affiliated with or endorsed by Meta, and connects through Meta's public Marketing API on its own terms. Both are legitimate choices, but they carry different trust, support and data-handling implications. Always review the source and terms before connecting a production ad account.

If you manage Meta advertising campaigns, you know the drill. Export performance data. Open a spreadsheet. Manually calculate ROAS by ad set. Stare at frequency numbers trying to judge whether creative fatigue is setting in. Shift budgets. Wait three days to see if it worked. Repeat.

It's not that this process doesn't work — it does. It's that it's slow, reactive, and completely dependent on when you have time to look at the data. By the time you catch a creative burning out or a budget leaking into a non-converting audience, you've already lost money.

That's changing. A combination of Claude AI and the Meta Ads MCP server is now giving marketers — from solo operators to agency teams managing millions in spend — something genuinely new: an AI assistant that has live access to your campaign data and can act on it through natural conversation.

Table of Contents

1.  [The Official Meta MCP Server](#official-server)
2.  [What Is the Meta Ads MCP Server?](#what-is-meta-ads-mcp)
3.  [25 Tools Available](#tools-available)
4.  [Setup Options](#setup-options)
5.  [Real Workflows](#real-workflows)
6.  [Verified Results](#verified-results)
7.  [Limitations](#limitations)
8.  [Combining MCP Servers](#combining-servers)
9.  [Getting Started](#getting-started)

## The Official Meta Ads MCP Server — What Changed in 2026

For most of this guide's life, connecting Claude to Meta Ads meant picking a community server, registering your own Meta Developer App, and then waiting. App Review took four to six weeks, and it was the single most common point at which people gave up. Plenty of marketers read a guide like this one, hit the review queue, and never came back.

That barrier is gone. On **29 April 2026**, Meta shipped a first-party hosted MCP server:

```
https://mcp.facebook.com/ads
```

It is a standard Streamable HTTP MCP server, so any compliant client can use it — Claude, Claude Code, Cursor, ChatGPT Connectors, Codex, Windsurf, Gemini, or your own. There is nothing to install and nothing to host.

### How the authentication actually works

Point an MCP client at that URL unauthenticated and Meta answers with a textbook OAuth challenge — a `401` carrying a `WWW-Authenticate` header that tells your client exactly where to go and what to ask for:

```
WWW-Authenticate: Bearer
  resource_metadata="https://mcp.facebook.com/.well-known/oauth-protected-resource/ads",
  scope="ads_management ads_read catalog_management
         business_management pages_show_list
         instagram_basic ads_mcp_management"
```

Those seven scopes are worth reading closely, because they tell you precisely how much access you are handing over:

-   `ads_read` — read campaigns, ad sets, creatives and insights. This alone covers every reporting workflow in this guide.
-   `ads_management` — create and modify campaigns, budgets and targeting. This is the one that lets an agent _spend money_.
-   `business_management` — read and manage Business Manager assets.
-   `catalog_management` — product catalogs, for commerce and dynamic ads.
-   `pages_show_list` and `instagram_basic` — the Pages and Instagram accounts your ads run from.
-   `ads_mcp_management` — Meta's own scope governing MCP access itself.

The practical advice: if you are starting out, grant read scopes and live in reporting for a couple of weeks before you hand any agent `ads_management`. The workflows in section 4 that deliver most of the value — audience overlap, frequency diagnosis, creative fatigue detection — are all read-only. The write scopes are where a confidently wrong model gets expensive.

🔍 Check what you are about to connect

Whether you use Meta's server or a community one, it is worth knowing what a server exposes before you authorise it against a live ad account. Our free [MCP Security Scanner](/mcp-security-scanner) runs 35+ checks — transport, auth, protocol compliance, injection surface, information disclosure — and grades the result A–F. Paste a URL, no sign-up.

### Official or community — which should you actually use?

The official server is now the right default for most people: no App Review, no self-hosting, first-party support, and OAuth scoping you can reason about. Choose a community server when you need something it does not give you — a self-hosted deployment for compliance reasons, multi-platform coverage behind one interface, or a specific tool the official surface does not expose yet. Both paths are covered below.

One thing worth saying plainly: the whole ad-platform landscape moved in the same ten-week window. Google Ads shipped an official (self-hosted) MCP on 28 April, Meta on 29 April, TikTok announced theirs at TikTok World on 13 May, and X launched [a hosted server at `api.x.com/mcp`](/blog/x-twitter-ads-mcp-claude-automation-guide) on 30 June. LinkedIn and Reddit still have no first-party server, so for those the third-party route in this guide remains the only route.

## What Is the Meta Ads MCP Server?

The [Meta Ads MCP server](https://github.com/pipeboard-co/meta-ads-mcp) is an open-source Model Context Protocol server built by Pipeboard that bridges Claude AI directly to the Meta Marketing API. It predates Meta's official endpoint and remains the most mature community option — worth understanding both because many teams already run it and because its tool design is a good map of what any Meta Ads server needs to cover.

Instead of Claude reasoning about your campaigns from a screenshot or a pasted CSV, it has a live, authenticated connection to your actual ad account data — campaigns, ad sets, creatives, audiences, insights, and budget schedules. All of it, in real time.

📊 Live Data Access

Pull real-time CTR, ROAS, CPC, Frequency — no CSV exports

🎯 Full Campaign Control

Create, update, and pause campaigns, ad sets, and creatives

🧠 Audience Intelligence

Search, validate, and recommend interest targeting options

💬 Natural Language

No dashboards, no exports — just ask Claude what you need

## The 25 Tools Available — What Claude Can Actually Do

The meta-ads-mcp server exposes 25 tools across six functional areas. Here is exactly what each covers:

### Account & Campaign Management

Tool

What It Does

`get_ad_accounts`

List all ad accounts your token has access to

`get_campaigns`

Fetch all campaigns with optional status filtering

`create_campaign`

Launch a new campaign with objective and budget

`get_campaign_details`

Pull full details on any specific campaign

### Ad Set Operations

Tool

What It Does

`get_adsets`

Retrieve ad sets, optionally filtered by campaign

`create_adset`

Create a new ad set with targeting and bidding

`update_adset`

Modify frequency caps, bid strategy, schedule

`get_adset_details`

Deep-dive into any ad set's full configuration

### Creative Management

Tool

What It Does

`get_ad_creatives`

Retrieve creative details for any ad

`create_ad_creative`

Build new creatives with image hash and copy

`update_ad_creative`

Revise existing creative content

`upload_ad_image`

Upload images directly for use in ads

`get_ad_image`

Download and visually inspect a live ad image

### Performance Analytics

Tool

What It Does

`get_insights`

Pull performance data for any account, campaign, ad set, or individual ad with custom date ranges — CTR, ROAS, CPC, CPM, Frequency, and more

### Audience & Targeting

Tool

What It Does

`search_interests`

Find interest targeting options by keyword

`get_interest_suggestions`

Get recommendations based on existing selections

`validate_interests`

Confirm interest names or IDs are valid before use

`search_behaviors`

Access Meta's behavior targeting specifications

`search_demographics`

Retrieve demographic targeting options

`search_geo_locations`

Find geographic targets by country, city, or region

## How to Connect It: Four Setup Options

### Option 1 — Meta's Official Hosted Server (Recommended)

**Setup time: ~5 minutes  |  Cost: Free (you pay for ads, not the server)**  
First-party, no App Review, no infrastructure. The right default for almost everyone since April 2026.

Point any MCP client at Meta's own endpoint and complete the OAuth flow in the browser. There is no package to install, no token to paste into a config file by hand, and no Developer App to register.

```
{
 "mcpServers": {
 "meta-ads": {
 "type": "http",
 "url": "https://mcp.facebook.com/ads"
 }
 }
}
```

Your client will hit the `401`, follow the `resource_metadata` pointer, walk you through Meta's Business OAuth consent screen, and store the resulting token itself. Grant the narrowest scope set that covers what you actually want to do — `ads_read` is enough for every reporting workflow below.

**Prefer not to touch a config file at all?** Paste the same URL into [MCP Agent Studio](/mcp-agent-studio) and it runs in your browser against 60+ models, with every tool call and its full JSON visible as it happens. It is the fastest way to find out whether this is useful to you before you commit to a desktop setup.

### Option 2 — Remote MCP via Pipeboard

**Setup time: ~5 minutes  |  Cost: $49/month**  
Best for marketers, agencies, and non-technical teams who want it running today.

[Pipeboard](https://pipeboard.co) hosts and manages the MCP server for you. Authorize your Meta Business Manager, get a server URL, and paste it into Claude Desktop's config. No Node.js, no terminal, no infrastructure.

### Option 3 — Local Installation (Free, Open Source)

**Setup time: 30–60 minutes  |  Cost: Free**  
Best for developers and technical marketers who want full control.

```
# Prerequisites: Node.js 18+
git clone https://github.com/pipeboard-co/meta-ads-mcp
cd meta-ads-mcp
npm install

# Authenticate with Meta
bash meta_ads_auth.sh
```

Then add to your Claude Desktop config at `~/.config/claude/claude_desktop_config.json`:

```
{
 "mcpServers": {
 "meta-ads": {
 "command": "node",
 "args": ["/path/to/meta-ads-mcp/index.js"],
 "env": {
 "META_ACCESS_TOKEN": "your_access_token_here"
 }
 }
 }
}
```

### Option 4 — Composio (Managed, Developer-Friendly)

**Setup time: ~10 minutes  |  Cost: Usage-based**  
Good middle ground between fully managed and fully self-hosted.

[Composio](https://composio.dev/toolkits/metaads/framework/claude-code) provides a managed Meta Ads MCP integration with pre-built authentication — useful if you're already using Composio for other tool integrations.

**Before committing to any setup:** You can explore how MCP servers work and what tools they expose at [MCP Playground](/mcp-test-server) — paste any MCP server URL and inspect its tools in your browser, no install needed.

## Real Workflows: What This Actually Looks Like

### Workflow 1 — Morning Campaign Audit (What Used to Take 2 Hours)

The old way

-   Log into Meta Ads Manager
-   Export data for each campaign
-   Open spreadsheet, calculate ROAS
-   Identify outliers manually
-   Implement changes in Meta
-   **Total: 90–120 minutes**

The new way

-   Open Claude Desktop
-   Ask for a full performance summary
-   Claude pulls live data via MCP
-   Review flagged issues
-   Approve recommended actions
-   **Total: ~45 seconds**

Here is what the actual conversation looks like:

```
You: "Give me a full performance summary of all active campaigns
for the last 7 days. Flag anything with ROAS below 2.0 or
frequency above 3.5."

Claude pulls live data via get_insights and get_campaigns, then responds:

"Here's your summary across 8 active campaigns:

🔴 NEEDS ATTENTION:
- 'Retargeting — Warm Audience' — Frequency: 4.2, ROAS: 1.4
 Recommendation: Pause or refresh creative. Audience is fatigued.

- 'Lookalike 5% — Purchase' — ROAS: 1.7, CPC up 34% week-over-week
 Recommendation: Review creative. Consider tightening audience.

✅ PERFORMING WELL:
- 'TOF — Interest Stack A' — ROAS: 4.1, Frequency: 1.8
 Recommendation: Budget increase candidate.

- 'TOF — Broad Auto' — ROAS: 3.6, CPA: $18.40 (target: $22)
 Recommendation: Maintain. Room to scale."
```

### Workflow 2 — Creative Fatigue Detection Before You Lose Money

Creative fatigue is the silent budget killer in Meta campaigns. Most advertisers only notice it when ROAS has already dropped significantly. With Claude connected to live Meta data, you can catch it on the way down — not after.

A DTC fitness brand was running a top-performing ad at $180/day. Historical data showed their creatives typically peaked around day 8–10 before CTR began decaying. With the Meta Ads MCP integration, Claude monitored CTR trend daily. On day 6, CTR had dropped 18% from its peak — within normal variance, but at an earlier-than-expected rate.

Claude flagged it: _"CTR trending down faster than your historical baseline for this placement. Recommend preparing creative refresh before frequency reaches 2.8."_

The brand launched a refreshed angle on day 7. Result: maintained 3.2x ROAS through day 14, instead of the typical decay to 1.8x they'd previously experienced.

```
You: "Check CTR trend for all ads in the 'TOF Prospecting'
campaign over the last 14 days. Flag any where CTR has dropped
more than 15% from its 3-day peak, and show current frequency."
```

### Workflow 3 — Budget Reallocation Based on Live ROAS

A SaaS company was running $3,000/day split evenly across 5 campaigns — $600 per campaign. Claude's analysis via `get_insights` revealed that 2 campaigns were delivering 80% of all conversions at 40% below target CPA, while 3 campaigns were consistently running 60% above target with no improving trend.

The reallocation Claude recommended:

Campaign

Before

After

Reason

TOF — Broad

$600

$900

Strong ROAS, not spend-constrained

Retargeting — Purchasers

$600

$750

Highest ROAS of all campaigns

Lookalike — 1%

$600

$600

Stable — maintain

Interest Stack B

$600

$450

Above target CPA, limit exposure

Interest Stack C

$600

$300

Consistently underperforming

**Outcome:** Monthly qualified demo bookings went from 147 to 231 — with identical total spend. No new creative. No new audiences. Just smarter money distribution backed by live data.

### Workflow 4 — Audience Discovery and Interest Validation

Claude can search and validate Meta's entire interest graph before you waste budget on poorly matched audiences:

```
You: "I'm launching a campaign for a B2B project management tool
targeting SMB owners. Suggest 10 interest targeting options,
validate they exist in Meta's system, and estimate audience size."

Claude calls search_interests, validate_interests, get_interest_suggestions:

"Here are 10 validated interests for your campaign:

1. Small Business (180M) — Broad but relevant
2. Entrepreneurship (87M) — Strong intent signal
3. Project Management (23M) — High relevance, smaller pool
4. Asana (4.2M) — Competitor targeting, strong intent
5. Trello (6.8M) — Competitor targeting
6. Monday.com (3.1M) — Competitor targeting, high intent
7. Business productivity (41M) — Medium relevance
8. SaaS (8.3M) — B2B signal
9. Remote work (67M) — Post-2020 behavioral signal
10. Startup (52M) — Overlaps with entrepreneurship

Recommended approach: Start with interests 3, 4, 5, 6 stacked.
Total addressable audience ~37M, all showing commercial intent
signals directly relevant to your offer."
```

## Verified Results: What Teams Are Actually Reporting

**Important:** Every result below is sourced from a verified, published case study or official research. We have not included speculative projections.

90%

Reduction in operational work time

Advolve — Anthropic Case Study

15%

Increase in customer ROAS

Advolve — Anthropic Case Study

8→20

Client accounts, same 3-person team

Agency case study — Pipeboard

$3,200

Monthly waste found in 10 minutes

E-commerce brand — Pipeboard

The most credible data point comes directly from [Anthropic's official customer page for Advolve](https://claude.com/customers/advolve) — a company that uses Claude to automate digital marketing workflows at scale. Their published results: 90% reduction in operational work time, and a 15% increase in ROAS for their managed clients. These are not projections or estimates — they are published outcomes from a live production deployment.

## What Claude Cannot Do — Be Realistic

This integration is powerful — but it has real limits:

-   Cannot override Meta's ad policies or bypass platform restrictions
-   Cannot guarantee ROAS improvements — offer quality and landing pages still matter
-   Cannot replace human judgment on brand strategy and creative direction
-   Cannot access TikTok, Google Ads, or other platforms without separate MCP servers
-   Cannot take fully autonomous action without your review unless explicitly configured

The correct mental model: **Claude is an extremely capable analyst and operator who works at AI speed, but you are still the strategist.** The best results come from teams who use Claude to handle data-heavy operational work while humans focus on strategy, creative, and positioning.

## Combining Meta Ads MCP with Other Servers

The real power compounds when you chain Meta Ads MCP with other data sources:

Full-Funnel Attribution

**Meta Ads MCP + Stripe MCP + PostgreSQL MCP**  
Claude sees ad spend, revenue generated, and customer LTV in one view. Calculates true blended ROAS accounting for subscription revenue, not just first purchase.

Creative → Production Automation

**Meta Ads MCP + Slack MCP + Notion MCP**  
Claude detects underperforming creative → Posts alert to #paid-media Slack channel → Creates a Notion brief for the creative team with performance context. Total human hours: zero.

Automated Weekly Reporting

**Meta Ads MCP + Google Sheets MCP**  
Claude pulls 7-day performance data across all accounts, formats it into your standard reporting template in Google Sheets. What used to take 3 hours now takes 4 minutes.

## Getting Started Today

If you are a marketer (non-technical):

1.  Open [MCP Agent Studio](/mcp-agent-studio) and paste `https://mcp.facebook.com/ads`
2.  Complete Meta's OAuth consent — grant `ads_read` only, to begin with
3.  Pick a model and run the morning audit prompt from section 4
4.  Once it earns its place, move it into Claude Desktop with the Option 1 config

If you are a developer:

1.  Add `https://mcp.facebook.com/ads` to your client as an `http` server
2.  Scan it first with the [Security Scanner](/mcp-security-scanner) if it is going near production
3.  Need self-hosting or multi-platform? Clone [pipeboard-co/meta-ads-mcp](https://github.com/pipeboard-co/meta-ads-mcp) instead
4.  Building your own ads server? Run [evals](/mcp-evals) to prove an agent can actually use your tools

Want to explore MCP before committing to a setup?

Connect to any MCP server in your browser — inspect tools, run live requests, no install needed.

[Open MCP Playground →](/mcp-test-server) [Browse MCP Servers Directory](/mcps)

**Disclaimer:** `mcp.facebook.com/ads` is operated by Meta Platforms, Inc. The community servers covered here — including pipeboard-co/meta-ads-mcp — are unofficial and are not affiliated with or endorsed by Meta. Whichever you choose, test against a staging account before connecting production ad budgets, and grant write scopes only once you trust the workflow.

## Frequently Asked Questions

**Does Meta have an official MCP server?+**

Yes, since 29 April 2026. It is a hosted Streamable HTTP endpoint at `https://mcp.facebook.com/ads`, operated by Meta and authenticated with standard Business OAuth. You no longer need to register a Meta Developer App or wait through App Review, which was the biggest drop-off point in the old setup. Community servers like pipeboard-co/meta-ads-mcp still exist and are still useful — mainly for self-hosting and multi-platform coverage — but the official endpoint is the right default for most people.

**What permissions does the official Meta MCP server ask for?+**

Its OAuth challenge requests seven scopes: `ads_management`, `ads_read`, `catalog_management`, `business_management`, `pages_show_list`, `instagram_basic` and `ads_mcp_management`. The one to think hardest about is `ads_management` — that is what allows an agent to create campaigns and move budgets, i.e. to spend money. Every reporting and diagnostic workflow in this guide runs on `ads_read` alone, so start there and add write access only once you trust the workflow.

**Does this work with Instagram ads too?+**

Yes. Instagram ads are managed through the same Meta Ads Manager and Meta Marketing API. The MCP server gives Claude access to campaigns running on Facebook, Instagram, Messenger, and the Audience Network through a single connection — the official server requests the `instagram_basic` scope precisely so it can see the Instagram accounts your ads run from.

**Is my ad account data safe?+**

For the self-hosted open-source option, your Meta access token stays on your machine and never passes through a third party. For managed services like Pipeboard, review their privacy policy before connecting production accounts. As a best practice, use a token with the minimum required permissions and create a dedicated system user in Meta Business Manager.

**Can Claude automatically make changes to my campaigns without my approval?+**

Only if you explicitly instruct it to. By default, Claude will surface recommendations and wait for your confirmation before taking any action. You are always in control of what gets executed.

**Does this replace Meta Ads Manager?+**

No — it complements it. Claude through MCP is best for analysis, optimization decisions, and bulk operations. Meta Ads Manager remains your source of truth for creative uploads, billing, and detailed reporting dashboards. Most teams use both: Claude for the operational layer, Meta Ads Manager for final review.

**Can I use this with Google Ads too?+**

Google Ads requires a separate MCP server — the Meta endpoint only connects to Meta's Marketing API. Google shipped its own official Ads MCP on 28 April 2026, one day before Meta's, though Google's is self-hosted rather than a hosted endpoint you can simply point a client at. TikTok announced an official server at TikTok World on 13 May 2026, and X launched a hosted one at `api.x.com/mcp` on 30 June. LinkedIn and Reddit still have no first-party server. Browse what is available in our [MCP Server Registry](/mcp-registry).

**How do I know an ads MCP server actually works before I trust it with a budget?+**

Test it, rather than assuming. Paste the server URL into [MCP Agent Studio](/mcp-agent-studio) and watch which tools a model picks and what arguments it builds — that alone catches most surprises. If you are building or operating an ads server yourself, run [MCP evals](/mcp-evals): they generate a test suite from your real tool schemas, grade protocol behaviour deterministically in code, and use one model to drive realistic tasks while another grades whether the answers were genuinely supported by the tool output. Before connecting anything to a live account, run the [Security Scanner](/mcp-security-scanner) over it.

## Related Guides

-   [What Is the Model Context Protocol (MCP)? A Developer's Guide](/blog/what-is-model-context-protocol)
-   [MCP vs REST API — What's Different?](/blog/mcp-vs-rest-api-whats-different)
-   [Build a Custom MCP Server Using Node.js](/blog/build-custom-mcp-server-using-nodejs-simple-guide)
-   [Why Security Matters in MCP Servers](/blog/mcp-security-best-practices)
-   [MCP Agent Studio — Chat with Your MCP Server Using AI](/blog/mcp-agent-studio-guide)

🤖 Test your Meta Ads MCP server with a live AI agent

Once your server is running, use [MCP Agent Studio](/mcp-agent-studio) to chat with it directly — ask for campaign summaries, ad performance, and budget recommendations in plain English. Works with Claude, GPT, Gemini and more. Free credits on sign-up.

---

_Canonical page: https://mcpplaygroundonline.com/blog/automate-meta-ads-claude-ai-mcp — MCP Playground (mcpplaygroundonline.com), the free browser-based tool for testing MCP servers and building AI agents._