← Back to API Documentation

MCP Server

The Fanpage Karma MCP (Model Context Protocol) Server lets AI assistants and automation platforms access your social media analytics data directly through the Data API v2.

MCP is an open standard that enables AI assistants to use external tools. Instead of copying data manually, you can ask your AI assistant to fetch metrics, posts, and time series data from your Fanpage Karma account in natural language.

How It Works

1. Connect your MCP client to the server URL (see setup guides below).
2. Authenticate — either with your API token or via OAuth login.
3. Ask questions in natural language — the assistant calls the right tool and returns the data.

Server URL

https://app.fanpagekarma.com/api/v2/mcp

Works with all MCP clients that support Streamable HTTP.

Prerequisites & Pricing

The MCP Server uses the Fanpage Karma Data API v2. To use it, you need an active API subscription (included in Gold plans and above).

API usage is billed per profile queried

Each profile you query via the MCP Server counts towards your API usage and is billed per profile per month (not per request). Once a profile is queried in a billing period, all further requests for that profile are included at no extra cost.

See API Documentation → Prerequisites & Pricing for full pricing details, including Basic and Enterprise plans.

Use the list_queried_profiles tool to check which profiles have been queried in the current billing period. See Settings → API for your current usage.

Authentication

The MCP Server supports two authentication methods. Choose the one that fits your client:

Bearer Token

For desktop apps, CLI tools, and automation platforms. You configure a static API token in the client.

Getting Your API Token

1. Log in to Fanpage Karma.
2. Go to Settings → API.
3. Copy your API token.

The token is sent as Authorization: Bearer YOUR_TOKEN header with every request. See the setup guides below for client-specific configuration.

OAuth 2.1

For browser-based clients that support the OAuth 2.1 standard (e.g. Claude Web). No API token is needed — you authorize access by logging in to your Fanpage Karma account.

The client handles the OAuth flow automatically (dynamic client registration, PKCE, token refresh). See Browser-based Clients (OAuth) for details.

Compatibility

The MCP Server works with any client that supports the MCP standard. Depending on the client, authentication is done via Bearer Token or OAuth 2.1.

Client Auth Method Notes
Claude Desktop Bearer Token Via claude_desktop_config.json
Claude Code (CLI) Bearer Token Via claude mcp add or .mcp.json
Cursor / Windsurf Bearer Token Custom header configuration
n8n Bearer Token MCP Client Tool node
ChatGPT Desktop Bearer Token Custom header configuration
Claude Web (claude.ai) OAuth 2.1 Add as Connector in Settings → Connectors
Other browser-based clients OAuth 2.1 Any MCP client supporting OAuth 2.1 with PKCE

Setup: Claude Desktop

Add the following to your Claude Desktop configuration file:

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json

{
"mcpServers": {
"fanpagekarma": {
"url": "https://app.fanpagekarma.com/api/v2/mcp",
"headers": {
"Authorization": "Bearer your-api-token-here"
}
}
}
}

Restart Claude Desktop after saving the file.

Setup: Claude Code

Run the following command in your terminal:

claude mcp add --transport http fanpagekarma https://app.fanpagekarma.com/api/v2/mcp \
--header "Authorization: Bearer your-api-token-here"

Or add to your project's .mcp.json:

{
"mcpServers": {
"fanpagekarma": {
"type": "http",
"url": "https://app.fanpagekarma.com/api/v2/mcp",
"headers": {
"Authorization": "Bearer your-api-token-here"
}
}
}
}

Setup: n8n

Use the MCP Client Tool node with the hosted service:

Step 1: Add Node

Add an MCP Client Tool node to your workflow.

Step 2: Configure Authentication

Authentication: Bearer Token
Token: Your Fanpage Karma API token

Step 3: Set Endpoint

Streamable HTTP (recommended):
https://app.fanpagekarma.com/api/v2/mcp

SSE (legacy):
https://app.fanpagekarma.com/api/v2/mcp/sse

Browser-based Clients (OAuth)

MCP clients that run in the browser can connect via OAuth 2.1 instead of a static API token. You authorize access by logging in to your Fanpage Karma account — no token copy-paste needed.

The general setup for any OAuth-capable MCP client:

Step 1: Add the MCP Server

In your client's settings, add a new MCP server / connector and enter the server URL:
https://app.fanpagekarma.com/api/v2/mcp

Step 2: Authorize

The client will redirect you to the Fanpage Karma login page. After logging in, you will see a consent screen showing the permissions being requested. Click Authorize to grant access.

Step 3: Start Using

Once authorized, you can use all Fanpage Karma tools directly in your conversations. Ask questions in natural language just like with any other MCP setup.

Example: Claude Web (claude.ai)

In claude.ai, go to Settings → Connectors, click Add Custom Connector, and enter the server URL above. Claude handles the rest automatically.

How It Works Under the Hood

The OAuth 2.1 flow is handled automatically by your client (dynamic client registration, PKCE, token refresh). Your access is tied to your Fanpage Karma account and requires an active API subscription. You can revoke access at any time by removing the server from your client settings.

Other Clients

Any MCP-compatible client (Cursor, Windsurf, ChatGPT, custom integrations, etc.) can connect using the HTTP endpoints below. See Authentication for how to authenticate.

Bearer Token Header

For clients using Bearer Token authentication, include the token in every request:

Authorization: Bearer your-api-token-here

Endpoints

Method Endpoint Description
POST /api/v2/mcp Send messages (Streamable HTTP)
GET /api/v2/mcp SSE stream (requires mcp-session-id header)
DELETE /api/v2/mcp Close session
GET /api/v2/mcp/sse Establish SSE connection (legacy)
POST /api/v2/mcp/messages Send messages via SSE (legacy)

Most clients support Streamable HTTP. Use the SSE endpoints only if your client requires the legacy SSE transport.

Available Tools

The MCP server exposes seven tools that map to the API v2 endpoints:

Tool Description Parameters
list_connected_profiles List your connected social media profiles with extended analytics
list_queried_profiles List profiles queried via API in current billing period
list_available_metrics List available metric keys, labels and types for a network and endpoint networkrequired, endpointrequired
get_profile_metrics Get profile-level metrics (followers, engagement, reach, etc.) networkrequired, profile_idrequired, from, to, metrics
get_profile_posts Get posts with engagement metrics networkrequired, profile_idrequired, from, to, limit, offset
get_profile_stories Get Instagram Stories data profile_idrequired, from, to, limit, offset
get_profile_daily Get daily time series for specific metrics networkrequired, profile_idrequired, metricsrequired, from, to

Date parameters use YYYY-MM-DD format. If omitted, the last 28 days are used.
Use list_available_metrics or GET /api/v2/{network}/{endpoint}/metrics to discover available metric keys.
The get_profile_daily tool uses the /profile/daily endpoint path.

Supported Networks

The following networks are available for all profile tools:

  • facebook
  • instagram
  • youtube
  • linkedin
  • tiktok
  • threads
  • bluesky
  • pinterest

The get_profile_stories tool is only available for Instagram.

Usage Examples

Once configured, you can ask your AI assistant questions like:

Profile Discovery

“List all my social media profiles.”

Metrics Analysis

“Show me the metrics for my Instagram account @example from 2024-01-01 to 2024-01-31.”

Post Performance

“Get the top performing posts from my Facebook page this month.”

Cross-Network Comparison

“Compare engagement rates across all my profiles.”

Time Series Trends

“Show me the daily follower trend for my X profile over the last 30 days.”

Instagram Stories

“Show me my Instagram Stories performance from last week.”