The average enterprise on Netlify ships 140 times per week. Where does your team rank?

ContactSign Up
Community Plugin
View plugin on GitHub


A gatsby source plugin for pulling in third party api data.


  • Pulls data from configured api url
  • Uses custom name to allow for multiple instances of plugin
  • Option to download the json data to a configurable path
  • Option to only download the json data, and skip inserting it into GraphQL
  • Supports simple authentication through axios


npm install --save gatsby-source-apiserver

Migrate for Gatsby-v2 release:

Please checkout version 2.0.0 or next

npm install --save gatsby-source-apiserver@next

Change logs

  • 2.1.6:
    • Support pagination for recurring fetch big amount of data from server. See calculateNextPage in example configuration
  • 2.1.5:
    • Allow caching for slow API servers. See example configuration
    // enable disk caching
    allowCache: false,
    // if allowCache is true, then the cache will be purged after the
    // specified amount of time
    maxCacheDurationSeconds: 60 * 60 * 24,
  • 2.1.3:
    • Add ability to refresh node in development by ENABLE_GATSBY_REFRESH_ENDPOINT env. See more
  • 2.1.2:
    • Upgrade axios to 0.19.0 to fix DoS vulnerability
    • Fix BoundActionCreators depracation warning
  • 2.1.1:
    • Support multiple entities for multiple api servers, pls take a look at attribute entitiesArray
    • Add request params
    • Support Auth0
  • 2.0.0: Support gatsby-v2

How to use

// Place configuration options in your gatsby-config.js

plugins: [
    resolve: "gatsby-source-apiserver",
    options: {
      // Type prefix of entities from server
      typePrefix: "internal__",

      // The url, this should be the endpoint you are attempting to pull data from
      url: ``,

      method: "post",

      headers: {
        "Content-Type": "application/json"

      // Request body
      data: {},

      // Name of the data to be downloaded.  Will show in graphQL or be saved to a file
      // using this name. i.e. posts.json
      name: `posts`,

      // Nested level of entities in response object, example: `data.posts`
      entityLevel: `data.posts`,

      // Define schemaType to normalize blank values
      // example:
      // const postType = {
      //   id: 1,
      //   name: 'String',
      //   published: true,
      //   object: {a: 1, b: '2', c: false},
      //   array: [{a: 1, b: '2', c: false}]
      // }
      schemaType: postType,

      // Request parameters
      // Only available from version 2.1.0
      params: {
        per_page: 1

      // Simple authentication, optional
      auth: {
        username: "myusername",
        password: "supersecretpassword1234"
      // enable disk caching
      allowCache: false,
      // if allowCache is true, then the cache will be purged after the
      // specified amount of time
      maxCacheDurationSeconds: 60 * 60 * 24,

      // Advanced authentication for Auth0
      // Only available from version 2.1.0
      auth0Config: {
        method: "POST",
        url: "https://MyAuth0Domain/oauth/token",
        headers: { "content-type": "application/json" },
        data: {
          grant_type: "password",
          username: "myusername",
          password: "PassAWordHere",
          audience: "Auth0APIAudience",
          scope: "openid",
          client_id: "AUTH0_CLIENT_ID",
          client_secret: "AUTH0_SECRET"
        json: true

      // Optional payload key name if your api returns your payload in a different key
      // Default will use the full response from the http request of the url
      payloadKey: `body`,

      // Optionally save the JSON data to a file locally
      // Default is false
      localSave: false,

      //  Required folder path where the data should be saved if using localSave option
      //  This folder must already exist
      path: `${__dirname}/src/data/auth/`,

      // Optionally include some output when building
      // Default is false
      verboseOutput: true, // For debugging purposes

      // Optionally skip creating nodes in graphQL.  Use this if you only want
      // The data to be saved locally
      // Default is false
      skipCreateNode: false, // skip import to graphQL, only use if localSave is all you want

      // Optionally re-source data when it changes and
      // `gatsby develop` is running.
      // Requires `ENABLE_GATSBY_REFRESH_ENDPOINT=true`.
      // See
      // Default is false
      enableDevRefresh: true,

      // Optionally override key used to re-source data
      // when `gatsby develop` is running.
      // Requires `enableDevRefresh: true`.
      // See setting directly above this one.
      // See also
      // Default is `id`
      refreshId: `id`,

      // Pass an array containing any number of the entity configuration properties (except verbose, auth0Config),
      // any not specified are defaulted to the general properties that are specified
      // Only available from version 2.1.0
      entitiesArray: [
          url: ``,
          method: "post",
          headers: {
            "Content-Type": "application/json"
          name: `posts`,
          // optional paging
          calculateNextPage: (curUrl, response, context) => {
            let page = 2
            if ( {
              page = + 1
   = page
            const url = `${page}`
            const hasNext = >= 100
            return { url, hasNext }

How to query

Data will be available at the following points in GraphQL.

all<TypePrefix><Name> or <TypePrefix><Name> where TypePrefix and Name is replaced by the name entered in the configuration options.

Dummy Node

This plugin will automatically add a dummy node to initialize Gatsby Graphql Schema, in order to avoid GraphQL errors when some fields are missing.

The dummy node will have id: 'dummy' and you will probably want to exclude it from createPage():

    .filter(({ node }) => !== "dummy")
    .map(({ node }, index) => (
      <li key={index}>{}</li>

or filter it out from your GraphQL query:

query {
    internalAllPosts(filter: {id: {ne: "dummy"}}) {
        // ...

Note: make sure you pass option schemaType to make dummy node works.

Conflicting keys

Some of the returned keys may be transformed if they conflict with restricted keys used for GraphQL such as the following ['id', 'children', 'parent', 'fields', 'internal']

These conflicting keys will now show up as alternative_id

© 2023 Gatsby, Inc.