What is MCP in social media scheduling?
MCP, the Model Context Protocol, is a standard way for AI assistants such as ChatGPT and Claude to use another app. In social media scheduling, it lets an assistant create, edit and schedule posts inside your scheduler.
Why it is different from an API key
With MCP and OAuth, you approve access in a sign-in screen. There is no key to copy, store or rotate, and you can revoke the connection at any time from the app. Tools that do not support OAuth, such as some automation platforms, can still use an API key.
What an assistant can do over MCP
In a scheduler that supports MCP, the assistant can list your brands and connected accounts, check what each network supports, look up the best times to post, search your media library or stock photos, write a draft and propose a publish time. In Flomixy every post the assistant creates goes through the normal approval flow before it can publish.
How to connect Claude
In Claude, open Settings, then Connectors, and add a custom connector with the server URL https://mcp.flomixy.com/mcp. Claude opens the Flomixy sign-in screen; log in, choose the workspace and approve access. After that you can ask Claude to draft a post for a brand and schedule it for a free slot.
How to connect ChatGPT
In ChatGPT, open Settings, then Apps and Connectors, and add a new connector with the same URL, https://mcp.flomixy.com/mcp. Sign in to Flomixy when asked and approve access. ChatGPT can then read your brands and accounts and create drafts that wait for approval.
How to connect Cursor or n8n
Tools without an OAuth flow connect with an API key. Create a key in the Flomixy workspace settings, then add the server URL https://mcp.flomixy.com/mcp to the tool's MCP configuration with the key as a bearer token. Treat the key like a password and revoke it when you stop using the tool.
Keeping a human in control
Letting an AI post on its own is a real risk for an agency: one wrong caption on a client account costs more than the time saved. Look for a scheduler where AI-created posts cannot skip approval, where you can see who created a post, and where a client still signs off before anything goes live.
What to look for
Check which actions the assistant can take, whether its posts go through your normal approval flow, whether the connection uses OAuth or a long-lived key, and whether it works across all of your brands or only one account at a time.
