Glossary

MCP (Model Context Protocol)

MCP, the Model Context Protocol, is an open standard that lets an AI assistant connect to outside tools and data through a common interface, so it can read information and take actions in other apps.

How MCP works

MCP was introduced by Anthropic in late 2024 as an open protocol, and it has since been adopted by many AI clients, including Claude, ChatGPT, Cursor, VS Code and Gemini CLI. It has two sides:

  • An MCP client: the AI app you talk to.
  • An MCP server: a service that exposes "tools" the AI can call, each with a name, a description and parameters.

When you ask the AI to do something, it decides which tool to call, the server runs it, and the AI reads the result. A remote MCP server lives on the internet; you add its URL to your AI app and sign in, usually with OAuth, so you never paste a password into the chat.

MCP in social media scheduling

For a scheduler, MCP means your AI can work with your real calendar instead of producing text you copy and paste. With Postika, an AI connected over MCP can list your projects and channels, read what is already scheduled and create posts as drafts, pending approval, scheduled or published right away. Photos and videos cannot be sent over MCP, so for Instagram, TikTok, Pinterest and YouTube the AI leaves the post for you to add the media.

Many schedulers now offer an MCP server, with different scopes. Setup steps for each AI are in the MCP docs and the AI hub; a longer explanation is in MCP for social media.

Try Postika free for 7 days

Connect your networks and your AI in minutes. Cancel anytime.