nightwatch-openai-plugin

Nightwatch.js plugin which adds a reporter which uses the OpenAI platform to provide AI-assisted error analysis for test failures.

The reporter sends the error message to an HTTP API service which uses the OpenAI API to inspect the error and provide actionable feedback.

For more details, read the full tutorial on Using the OpenAI platform to analyse automated test failures, published on our blog.

Installation

The plugin can be installed via npm:

npm install nightwatch-openai-plugin

Then update your nightwatch.conf.js file to include the plugin:

module.exports = {
  // ...
  plugins: ['nightwatch-openai-plugin']
  // ...
}

Configuring the AI Analysis Service

The plugin uses an HTTP API service to interact with the OpenAI API. By default a service is provided for demo purposes, which is configured in the .env file in this project.

You can host your own service by cloning the openai-nightwatch-service repository and running it with your own OpenAI API key.

When running the openai-nightwatch-service, you need to define the NIGHTWATCH_ANALYSIS_SERVICE_URL environment variable to point to the service URL. You can also use .env files.

For example, assuming you have the service running at http://localhost:4001, you can create a .env file with the following content in the root of your Nightwatch project:

NIGHTWATCH_ANALYSIS_SERVICE_URL=http://localhost:4001/analyze-error

License

MIT