Product Release: GraphQL + MCP

GraphQL is a query language for APIs that lets you request exactly the data you need in a single call. Instead of making multiple requests to different endpoints (like with REST), you write one query that specifies which fields you want, and the API returns just those fields.

For business verification, this means you can fetch a company's details, owners, and addresses in one request instead of three separate calls.

MCP (Model Context Protocol) is an open standard that lets AI agents connect directly to external tools and data sources. Think of it as a way for AI applications like Claude or Cursor to "talk to" other services without requiring custom code.

With Middesk MCP, AI agents can verify businesses and access identity data automatically. The agent simply figures out how to call Middesk on its own.

No. Our REST API remains fully supported and is the right choice for most integrations. GraphQL and MCP are additional options for specific use cases:

• Use GraphQL if you need to fetch complex, nested data efficiently or are building AI workflows where token costs matter

• Use MCP if you're building AI agents that need to verify businesses autonomously

• Use REST for standard integrations and straightforward API calls

The REST API requires multiple sequential calls to retrieve related business data. For example, fetching a business's identity, principals, and addresses typically requires 3-5 separate requests, each returning a full payload.

The GraphQL API lets you request exactly the data you need in a single query. Fetch a business with nested relationships (people, addresses, formation details) in one request, receiving only the fields you specify. This reduces API calls and response sizes, particularly valuable for AI agents operating under token budgets.

Request access to the GraphQL API, currently available to a limited set of customers for early testing.

Once approved, make queries to https://api.middesk.com/graphql using your existing Middesk API credentials in the Authorization header as a Bearer token.

We recommend using a GraphQL client to introspect and explore the schema. Start with simple queries to fetch business data by ID, then add nested fields as needed. See our GraphQL API documentation for authentication details, example queries, and pagination information.

Request access to Middesk MCP, currently in early testing.

Once approved, you'll receive instructions for installing the MCP server and configuring it with your AI agent. MCP-compatible agents like Claude and Cursor can then discover and use Middesk's verification capabilities automatically.

The setup is typically faster than building a REST integration since the agent handles request construction and response parsing on its own.

Yes. Both APIs access the same business verification data, so you can choose which works best for each use case.

Use the REST API for simple, single-resource operations and GraphQL for complex queries requiring multiple related resources. Your API credentials work for both endpoints, and webhook behavior is consistent.

The GraphQL API is in early access, and we're refining the schema based on customer feedback. We're starting with core business attributes (name, address, registration details, key relationships) and will evolve as we learn from usage.

The schema is introspectable, so your code or AI agent can discover available fields dynamically. This lets your integration adapt as we enhance the API. We'll communicate significant changes, but GraphQL's flexibility means your queries continue working as we add new capabilities.