Social media MCP
MCP (Model Context Protocol) is a standard way for an AI client to call tools. A social media MCP layer exposes your social accounts as those tools. MySoMeData is a read-only social media data API and MCP layer for AI. It connects social accounts to AI applications and software, providing structured historical performance data that can be analyzed, queried, and used to build products. MySoMeData is the MCP server for that read-only social media data.
What is MCP?
MCP lets an AI application discover tools and call them with structured arguments, instead of you pasting exports into a chat. The client is still Claude, Cursor, or another MCP host. MySoMeData is the server that returns your social data. The protocol does not grant the model permission to publish.
Why use MCP for social media data?
You ask a question in the AI you already use. The client calls a tool, receives normalized JSON, and answers from that payload. You skip maintaining a separate prompt that says “here is a CSV.” The same data remains available over the API when the caller is your own code rather than an MCP client.
What can an AI client query?
The MySoMeData MCP server exposes four read tools. Each accepts an optional platform filter and a period: a rolling window, the last completed calendar month, or the last completed calendar week.
- Account performance summary: engagement, format averages, what changed versus the previous period, outliers, and a short list of next steps when the sample supports them.
- Historical trend metrics: day-by-day totals and a stored sample of posts, Reels, pins, or videos.
- Organic outliers: items above the usual volume for their cohort. Instagram uses reach, TikTok uses video views, and Pinterest uses pin impressions.
- Audience demographics: Instagram follower, reached, and engaged mixes when Meta returns them; a TikTok follower mix; Pinterest ads audience interests.
How does the MySoMeData MCP layer work?
You connect the social accounts in MySoMeData. You then point an MCP client at the streamable HTTP endpoint /api/mcp. The client calls tools with your authorization. Responses are the same normalized social media data the REST API serves from https://api.mysomedata.com/v1. Access stays read-only.
Who is MCP useful for?
MCP is useful when a person asks questions in an AI client, or when a developer wants those tools inside an IDE. It is a poor fit for a scheduled HTTP workflow. Use the Social media data API from n8n or your own product, and use MCP from Claude social media analytics or another MCP host.