MCP server interface to Facebook Ads, offering programmatic access to ads data, management features and insights.
https://github.com/gomarble-ai/facebook-ads-mcp-serverYou're deep in campaign analysis, bouncing between Facebook Ads Manager tabs and Claude Desktop, manually copying performance data just to get AI-powered insights. There's a better way.
Every time you need Facebook Ads data for analysis, you're paying a productivity tax:
The Facebook Ads MCP Server eliminates this friction entirely.
This MCP server connects your AI assistant directly to Facebook's Marketing API with 20+ specialized tools. Query campaign performance, analyze ad creative effectiveness, or audit account structures without leaving your AI chat interface.
Real Example: Instead of exporting last month's campaign data and asking Claude to analyze it, you simply ask: "Show me campaign performance for the last 30 days and identify which ad sets are underperforming." The server fetches live data and provides immediate analysis.
get_campaign_insights - Pull performance metrics with custom date ranges and breakdownsget_adset_insights - Analyze ad set performance across multiple dimensionsget_ad_insights - Deep-dive into individual ad performancefetch_pagination_url - Handle large datasets seamlesslylist_ad_accounts - Audit multiple ad accounts quicklyget_activities_by_adaccount - Track changes and optimization historyget_details_of_ad_account - Pull account-level configuration and limitsget_ad_creative_by_id - Examine creative assets and copyget_adsets_by_campaign - Map campaign structure for optimizationPerformance Audits: Ask your AI to compare this month's ROAS across all campaigns and flag any spending anomalies. The server pulls live data, and your AI provides actionable recommendations immediately.
Creative Testing Analysis: "Which ad creatives in my DTC campaign have CTRs above 2% and what elements do they share?" Get insights without manual data exports or spreadsheet gymnastics.
Budget Optimization: Real-time analysis of cost-per-acquisition trends across ad sets, with AI-powered recommendations for budget reallocation—all in one conversation.
Change History Investigations: When campaign performance drops unexpectedly, trace recent changes across campaigns and ad sets without clicking through Ads Manager interfaces.
One-Click Installation: Download their installer at gomarble.ai/mcp for automatic setup with token generation handled for you.
Manual Setup: Standard Python installation with MCP client configuration. The server handles Facebook API authentication—you just need to provide your access token in the config.
Smithery Integration: Install via Smithery for Claude Desktop:
npx -y @smithery/cli install @gomarble-ai/facebook-ads-mcp-server --client claude
Add this to your Claude Desktop config and restart:
{
"mcpServers": {
"fb-ads-mcp-server": {
"command": "python",
"args": [
"/path/to/server.py",
"--fb-token",
"YOUR_FACEBOOK_ACCESS_TOKEN"
]
}
}
}
Now every conversation with Claude has direct access to your Facebook Ads data. No imports, no exports, no context switching.
This isn't about automating Facebook Ads management—it's about eliminating the data access bottleneck that slows down every optimization decision. When you can query live performance data conversationally and get AI analysis instantly, you move from reactive campaign management to proactive optimization.
The 74 stars and active community around this tool suggest other developers have already discovered this productivity multiplier. The question is: how much time are you still losing to manual data exports?
Ready to eliminate the context-switching tax? The Facebook Ads MCP Server transforms how you interact with your advertising data—and your optimization workflow will never be the same.