A lightweight serverless backend that securely searches Meta (Facebook) advertising interests using the Meta Graph API, with native Claude MCP support.
Meta Interest Finder is a lightweight backend service built for developers, marketers, AI applications, and automation tools that need to search Meta advertising interests securely.
Instead of exposing your Meta Access Token inside frontend applications, Meta Interest Finder keeps all communication with the Meta Graph API on the server.
The project is optimized for Vercel Serverless Functions, making deployment fast, simple, and scalable.
It also includes native Model Context Protocol (MCP) support, allowing Claude Desktop and other MCP-compatible AI assistants to search Meta advertising interests directly during conversations.
Searching Meta advertising interests isn't as straightforward as making a single API request from the browser.
Developers often run into problems like:
- Exposing Meta Access Tokens
- CORS restrictions
- Complex Graph API requests
- Difficult deployment
- Repetitive backend setup
Meta Interest Finder solves these problems by providing a production-ready backend that can be deployed in minutes.
Search thousands of Meta advertising interests directly through a simple REST API.
Your Meta Access Token never leaves your backend.
Unlike frontend implementations, sensitive credentials remain protected using server-side environment variables.
Optimized for Vercel Serverless Functions.
No infrastructure management.
No VPS required.
Deploy in minutes.
Connect directly to Claude Desktop using the Model Context Protocol (MCP).
Claude can search Meta advertising interests during conversations without exposing your Meta credentials.
Simple JSON API that can be integrated into:
- SaaS products
- Internal tools
- Marketing platforms
- AI agents
- Browser extensions
- Desktop applications
Built using a minimal Express.js backend with very few dependencies.
Simple architecture.
Easy to extend.
Designed for low-latency interest searching while minimizing unnecessary processing.
Search Request
│
▼
┌─────────────────────────┐
│ Meta Interest Finder │
│ Express API │
└─────────────────────────┘
│
▼
Meta Graph API
│
▼
Interest Search Results
│
▼
JSON Response
One of the most powerful features of this project is native support for the Model Context Protocol (MCP).
This allows Claude Desktop to communicate directly with Meta Interest Finder as an external tool.
Instead of manually opening Meta Ads Manager to search interests, Claude can perform the search on your behalf and return structured results inside your conversation.
Example workflow:
User
│
▼
Claude Desktop
│
▼
Meta Interest Finder (MCP)
│
▼
Meta Graph API
│
▼
JSON Results
│
▼
Claude Response
Meta Interest Finder can be integrated into a wide variety of applications.
Build audience research tools powered by Meta interest search.
Allow AI agents to discover advertising interests dynamically using MCP.
Create internal marketing dashboards without exposing API credentials.
Add Meta interest search functionality directly into your own software.
Discover related interests while planning Meta advertising campaigns.
- 🔍 Search Meta advertising interests
- 🔒 Secure backend architecture
- ⚡ Vercel optimized
- 🤖 Claude MCP support
- 🌍 REST API
- 📦 Lightweight Express server
- 🚀 Fast deployment
- 🔐 Environment variable support
- 📈 Marketing tool ready
- 🧩 Easy integration
Deploy Meta Interest Finder in under 5 minutes.
git clone https://github.com/lab1207/meta-interest-finder.git
cd meta-interest-finder
npm installCreate your environment variable:
META_TOKEN=YOUR_META_GRAPH_ACCESS_TOKENStart the development server:
npm startThe API is now ready to accept requests.
Before running the project, ensure you have:
- 18+
- Node.js
- npm
- A Meta Developer Account
- A valid Meta Graph API Access Token
Clone the repository.
git clone https://github.com/lab1207/meta-interest-finder.gitMove into the project.
cd meta-interest-finderInstall dependencies.
npm installCreate an environment file.
META_TOKEN=YOUR_META_TOKENStart the server.
npm startMeta Interest Finder is optimized for Vercel Serverless Functions.
Deployment only takes a few minutes.
Login to Vercel.
Click
New Project
Select
meta-interest-finder
Leave the Framework Preset as:
Other
Click
Deploy
Navigate to
Project
↓
Settings
↓
Environment Variables
Create
| Variable | Value |
|---|---|
| META_TOKEN | Your Meta Graph API Token |
Save the configuration.
Redeploy the project.
Open your deployed URL.
Example
GET
https://your-project.vercel.app/api/interests?q=fitnessIf everything is configured correctly, the API will return matching Meta advertising interests.
The project uses environment variables to securely communicate with the Meta Graph API.
| Variable | Required | Description |
|---|---|---|
| META_TOKEN | ✅ | Meta Graph API Access Token |
Never expose this token inside frontend applications.
Always keep it on the server.
Your Meta Access Token should include the following permissions.
| Permission | Purpose |
|---|---|
| ads_read | Access advertising interest search |
| business_management | Access Business Manager resources |
| read_insights | Retrieve audience information |
| pages_show_list | Access associated Pages |
These permissions represent the minimum recommended configuration.
Meta Interest Finder exposes a lightweight REST API.
GET /api/interests| Name | Required | Description |
|---|---|---|
| q | ✅ | Search keyword |
| limit | ❌ | Maximum number of results |
GET /api/interests?q=fitnessGET /api/interests?q=crypto&limit=25{
"success": true,
"results": [
{
"id": "6003139266461",
"name": "Fitness",
"audience_size_min": 1000000,
"audience_size_max": 3500000
}
]
}{
"success": false,
"error": "Invalid Meta Access Token"
}Meta Interest Finder includes native support for the Model Context Protocol (MCP).
This allows Claude Desktop to access Meta advertising interest search directly through your backend.
Supported endpoints:
GET /mcp
POST /mcpAvailable Tool
search_facebook_interests
Example conversation
User
↓
Find interests related to coffee shops.
↓
Claude
↓
Calls Meta Interest Finder
↓
Meta Graph API
↓
Returns matching interests
↓
Claude formats the results.
No API keys are exposed to Claude.
Everything remains securely hosted on your own server.
Meta Interest Finder works well with:
- Claude Desktop (MCP)
- Custom AI Agents
- SaaS Platforms
- Internal Marketing Tools
- Browser Extensions
- Desktop Applications
- Marketing Dashboards
- Audience Research Platforms
Meta Interest Finder follows a backend-first architecture.
Frontend
↓
Meta Interest Finder
↓
Meta Graph API
This approach ensures:
- API credentials remain private.
- Frontend applications never receive sensitive tokens.
- Easy key rotation.
- Better security.
- Cleaner architecture.
The project is intentionally lightweight and easy to understand.
meta-interest-finder
│
├── server.js
│ Main Express server
│ REST API
│ MCP Server
│ Meta Graph API integration
│
├── index.html
│ Simple browser testing interface
│
├── package.json
│ Project metadata and dependencies
│
├── vercel.json
│ Vercel routing and deployment configuration
│
└── README.md
The application follows a simple backend-first architecture.
Client Application
│
▼
Meta Interest Finder API
│
Authentication & Validation
│
▼
Meta Graph API
│
▼
JSON Response
This architecture ensures that your Meta Access Token is never exposed to frontend applications.
Meta Interest Finder was designed with simplicity and speed in mind.
- ⚡ Lightweight Express server
- 🚀 Serverless deployment
- 📦 Minimal dependencies
- 🔒 Secure environment variables
- 🌍 REST API
- 🤖 MCP compatible
- ☁️ Vercel optimized
Build tools that discover related Meta advertising interests.
Allow Claude or other AI assistants to search advertising interests using MCP.
Integrate interest search directly into your SaaS.
Power browser extensions without exposing Meta credentials.
Give marketing teams secure access to advertising interests.
No.
The token remains securely stored on the backend using environment variables.
Yes.
The project includes native support for the Model Context Protocol (MCP).
Yes.
Although optimized for Vercel, the project can run on any Node.js-compatible hosting platform.
Yes.
Please review the included license before using the project commercially.
Future improvements may include:
- Improved search capabilities
- Better documentation
- Additional Meta Graph API endpoints
- Performance optimizations
- Enhanced MCP integration
This project is actively maintained, and the roadmap will evolve based on community feedback and future development.
Contributions are welcome!
Whether you're fixing bugs, improving documentation, or adding new features, your help is appreciated.
- Fork the repository
- Create a new branch
- Commit your changes
- Push your branch
- Open a Pull Request
If you encounter a bug or have a feature request, please open a GitHub Issue.
When reporting issues, include:
- Operating System
- Node.js Version
- Deployment Platform
- Steps to Reproduce
- Expected Behavior
- Actual Behavior
This project is released under the MIT License.
You are free to:
- Use
- Modify
- Distribute
- Commercially use
the software under the terms of the MIT License.
Special thanks to:
- Meta Graph API
- Vercel
- Express.js
- Node.js
- Anthropic MCP
for making projects like this possible.
Happy Building! 🚀