MCP Server

The Adeptia MCP Server connects AI assistants to the Adeptia platform using the Model Context Protocol (MCP). It is designed for use with AI clients such as Claude Desktop, Cursor, and other compatible AI IDEs.

What the MCP Server does

The server lists capabilities across several areas, and AI assistants can call them as part of a conversation.

Capabilities available to AI assistants

The MCP Server organizes platform capabilities into distinct areas that an AI assistant can draw on during a conversation.

Capability area	What your AI assistant can help with
Discovery	Finding and exploring existing integration assets on the platform
Validation	Checking configurations for correctness before they are used
Monitoring	Querying the operational state of your integrations
Knowledge	Searching platform knowledge to answer questions or surface relevant information
Business rules	Working with the business rules defined on the platform
Mappings	Accessing and reasoning about data mappings
Templates	Using pre-built integration templates as starting points
Data enrichment	Applying enrichment logic to data as part of an integration
Format conversion	Converting data between formats supported by the platform

Reference data

In addition to interactive capabilities, the MCP Server makes reference data available to AI assistants so they can give more informed, accurate responses. This includes a catalog of the pre-built connectors available on the Adeptia platform, which an AI assistant can consult when helping you plan or configure an integration.

Guided prompts

The MCP Server includes a library of guided prompts, which are pre-built templates for common tasks and use cases. When you invoke one of these prompts in your AI client, it provides a structured starting point for the conversation.

Prompts fall into two categories:

Prompt type	Purpose
Use-case prompts	Guided templates for specific integration scenarios
Monitoring prompts	Pre-formed queries for checking the operational state of your integrations