Facebook Ads MCP Server
Meta Ads performance data and management tools for your AI assistant.
Data last scanned last week · Reviewed last month
Overview
This Python-based MCP server facilitates programmatic access to Meta Ads data, allowing AI assistants to interact with ad management features. Users can configure the server locally to query account performance and reporting metrics directly through supported MCP clients. The server supports two authentication paths: providing a manual Meta User Access Token or utilizing a connection to GoMarble's infrastructure to generate tokens. The installation process includes a local setup script for dependency management.
Our verdict
With a high risk level and a security score of 45/100, users should exercise caution. Key security factors include the lack of editorial review for authentication methods and the fact that this is a community-maintained project rather than an official Meta release. The reliance on an external service (GoMarble) for token generation warrants careful consideration of your data handling policies before integration.
- No direct dependencies reduce the supply-chain surface area
- Active development with commits within the last 24 hours
- MIT licensed open-source codebase
- Provides clear documentation for local environment installation
- Community-maintained without official support from Meta
- Authentication methods have not undergone independent security review
- Read-only capabilities remain unverified
- Absence of a repository homepage link
Compatibility
No compatibility data yet.
Frequently asked questions
›What are the system requirements for running this server?
You need Python 3.10 or higher and all dependencies listed in the requirements.txt file.
›Do I need to manually obtain a Meta Developer Access Token?
No, while you can manually provide your own token, you can also use the installer provided by GoMarble, which connects to their server to create one on your behalf.
›Does GoMarble store my Meta Access Token?
No, GoMarble does not store your token. It is saved locally on your machine for use with the MCP server.
›How do I configure the server for use with MCP clients like Claude Desktop?
You must add the server to your client's configuration file, providing the path to your server.py and your Meta access token as an argument.
›Is it recommended to use a virtual environment?
Yes, creating and activating a virtual environment is recommended to help manage project dependencies cleanly.
Badge
Maintain this server? Add the live badge to your README.