JS Task API Examples: using app-keys

To use Yagna REST API, your application must identify itself by api-key. In our examples we utilize the try_golem key, which is automatically created if you have the YAGNA_AUTOCONF_APPKEY variable set in the terminal where you start the Yagna service.

The try_golem api-key is easy to read and remember, and useful for quickstarts and testing examples. However, normally you should rather use unique 32-char api-keys to better protect access to your requestor and its wallet.

Yagna service is installed and running with the try_golem app-key configured.

Create a project folder, initialize a Node.js project, and install libraries.

mkdir golem-example
cd golem-example
npm init
npm i @golem-sdk/task-executor
npm i @golem-sdk/pino-logger

Copy the code into the index.mjs file in the project folder and run:

node index.mjs

You can generate a unique api-key with the yagna command:

yagna app-key create requestor

In the command above requestor is the name of the key. The output is the 32-char app-key.

You can list available app-keys in your yagna instance with:

yagna app-key list

You can identify your key by name and copy the value from the key column.

Once you have created an app-key you can use it in your requestor script replacing insert-your-32-char-app-key-here with the actual key.

import { TaskExecutor } from "@golem-sdk/task-executor";
import { pinoPrettyLogger } from "@golem-sdk/pino-logger";

(async () => {
  const executor = await TaskExecutor.create({
    logger: pinoPrettyLogger(),
    // replace 'try_golem' with 'insert-your-32-char-app-key-here'
    api: { key: "try_golem" },
    demand: {
      workload: {
        imageTag: "golem/node:20-alpine",
      },
    },
    market: {
      rentHours: 0.5,
      pricing: {
        model: "linear",
        maxStartPrice: 0.5,
        maxCpuPerHourPrice: 1.0,
        maxEnvPerHourPrice: 0.5,
      },
    },
  });

  try {
    const result = await executor.run(async (exe) => (await exe.run("node -v")).stdout);
    console.log("Task result:", result);
  } catch (err) {
    console.error("An error occurred:", err);
  } finally {
    await executor.shutdown();
  }
})();

There is yet another option to set the apiKey. By default, the JS SDK will try to read the value from the YAGNA_APPKEY environment variable and set it as the apiKey.

To assign an app-key value to the YAGNA_APPKEY environment variable use this command:

export YAGNA_APPKEY=insert-your-32-char-app-key-here

and then remove the following line from the example script:

api: {
  key: 'insert-your-32-char-app-key-here'
}

When the Yagna service starts, if the environment variable YAGNA_AUTOCONF_APPKEY has been set, its value will be used to create an autoconfigured app-key. However, this app-key isn't permanent; it will vanish upon restarting the service unless the YAGNA_AUTOCONF_APPKEY variable is set again. While this approach is handy for tutorials and demonstrations, for regular use it's recommended to have a distinct app-key. This provides better management over access to your Yagna REST API.

export YAGNA_AUTOCONF_APPKEY=insert-your-desired-app-key-here