Schema-first mocking

When your GraphQL schema grows, writing every mock response by hand becomes repetitive. Schema-first mocking lets you generate realistic responses automatically from the schema type definitions — no manual stub data required. You can then override specific fields where you need exact values.

Setup

Install the required packages:

npm install --save-dev @graphql-tools/schema @graphql-tools/mock graphql

Generate Responses From The Schema

Define your schema, create a mocked version of it, and execute queries against it inside a Mock Config handler:

import { makeExecutableSchema } from '@graphql-tools/schema';
import { addMocksToSchema } from '@graphql-tools/mock';
import { graphql as executeGraphQL } from 'graphql';
import { graphql } from 'mock-config-server';

const typeDefs = `
  type User {
    id: ID!
    name: String!
    email: String!
    role: String!
  }

  type Query {
    user(id: ID!): User
    users: [User!]!
  }
`;

const schema = makeExecutableSchema({ typeDefs });
const mockedSchema = addMocksToSchema({ schema });

graphql.query('GetUsers', async () => {
  const result = await executeGraphQL({
    schema: mockedSchema,
    source: '{ users { id name email role } }'
  });

  return result;
});

The mocked schema generates plausible values for every scalar field — strings, IDs, booleans, and numbers — so you get a valid response without defining any data yourself.

Customizing Specific Fields

Override individual types or fields while letting the schema generate everything else:

import { makeExecutableSchema } from '@graphql-tools/schema';
import { addMocksToSchema } from '@graphql-tools/mock';

const schema = makeExecutableSchema({ typeDefs });

const mockedSchema = addMocksToSchema({
  schema,
  mocks: {
    User: () => ({
      name: 'John Doe',
      role: 'admin'
    }),
    ID: () => '1'
  }
});

Only the fields you specify are overridden. Everything else is still auto-generated.

Reflecting Variables In The Response

Combine schema-generated data with incoming variables to make responses feel realistic:

type GetUserVariables = { id: string };

graphql.query<{ body: { variables: GetUserVariables } }>('GetUser', async ({ entities }) => {
  const id = entities.variables?.id;

  const result = await executeGraphQL({
    schema: mockedSchema,
    source: 'query GetUser($id: ID!) { user(id: $id) { id name email } }',
    variableValues: { id }
  });

  return result;
});

When To Use Schema-First Mocking

Schema-first mocking is most useful when:

the schema is large and maintaining stubs for every field is impractical
you want responses to stay in sync with schema changes automatically
you need a working mock quickly and do not care about exact field values

For scenarios where the exact response data matters — specific test cases, known edge cases, error states — write explicit stubs or resolver functions as described in Stub response.