Glossary
MCP (Model Context Protocol) Server
A lightweight program that exposes a specific tool, data source or service to AI models over the MCP protocol. Ethora ships its own MCP server so AI agents can integrate directly with the platform.
General definition
An MCP server is a lightweight program that exposes a specific tool, data source or service to AI models using the Model Context Protocol. It acts as an intermediary, making external systems accessible to AI through standardized communication.
MCP servers operate through three primary capabilities:
- Tools — actions the AI can perform, such as sending messages or querying a database
- Resources — data the AI can read, such as file contents or API responses
- Prompts — pre-written templates for common tasks
Communication happens over standardized JSON-RPC transport — typically stdio for local servers, or HTTP with Server-Sent Events for remote ones.
The Ethora MCP server
Ethora offers its own MCP implementation so AI agents can integrate with the platform’s infrastructure. It provides Ethora API access (authentication, chat room management, file uploads, digital assets), XMPP integration for real-time messaging and presence, platform resources like user profiles, chat history, documents and analytics, and native bot deployment inside chat rooms.
It especially helps developers using AI-assisted IDEs to build their projects, by giving the model documentation access and a working understanding of the platform. Learn more in the AI SDK.
