Micro‑Federation

Use micro‑federation to compose multiple Axolotl modules in a single project or monorepo (or via npm packages). Individual schemas are merged into a supergraph that your server uses at runtime.

You can also run the project examples/yoga-federated to watch how micro federation works!

To enable it, configure axolotl.json like:

{
    "schema": "schema.graphql",
    "models": "src/models.ts",
    "federation":[
        {
            "schema":"src/todos/schema.graphql",
            "models":"src/todos/models.ts",
        },
        {
            "schema":"src/users/schema.graphql",
            "models":"src/users/models.ts",
        }
    ]
}

schema/models point to the supergraph schema and its generated models. Federation flow:

1. Generate models for each entry in federation.
2. Merge those schemas into the supergraph saved at schema.
3. Generate models for the supergraph.

To run federated axolotls - each of folders, users and todos in our case has to provide their own resolvers.ts files based on local models and Axolotl. Then inside the main resolvers.ts file we need to merge the subgraph resolvers:

import { mergeAxolotls } from '@aexol/axolotl-core';
import usersResolvers from '@/src/users/resolvers.js';
import todosResolvers from '@/src/todos/resolvers.js';
 
export default mergeAxolotls(usersResolvers, todosResolvers);

Use the merged resolvers as the adapter input, just like a single‑schema setup.