Back
MCP
Extend agent capabilities with custom tools and data sources
Overview
MCP (Model Context Protocol) is a standard way for AI systems to connect to external tools and data sources through an “MCP server.” Lorikeet supports MCP connections for tools like Notion and Linear, allowing Lorikeet to use the capabilities those MCP servers expose (for example, reading pages or creating issues) without your team having to build a custom integration from scratch.
What it is
MCP is the fastest way to connect Lorikeet to supported tools because you don’t need to manually set up API calls, define schemas, or maintain custom integration logic. It also reduces ongoing maintenance risk, since the integration follows a standard interface and can evolve as the connected tool’s MCP capabilities expand.
How to integrate
In the Lorikeet app, you’ll select the MCP integration you want (such as Notion or Linear) and connect it by completing the provider’s authorization steps. Once connected, Lorikeet makes the available MCP capabilities accessible inside workflows so you can use them to read information or take approved actions. Your admin completes the connection in Lorikeet and the provider; Lorikeet provides guidance and validates the setup end-to-end.
