This GitHub Action triggers one or more Endtest test executions, waits for all of them to finish, and exposes their results as workflow outputs.

It supports both API response shapes:

• A single execution hash, usually returned when the request uses suite.
• Multiple comma-separated execution hashes, usually returned when the request uses label and matches multiple test suites.

on: [push]

name: Endtest

jobs:
  test:
    name: Endtest functional tests
    runs-on: ubuntu-latest

    steps:
      - uses: actions/checkout@v4

      - name: Run tests with the Critical label
        id: endtest
        uses: endtest-technologies/github-run-tests-action@v1.10
        with:
          app_id: ${{ secrets.ENDTEST_APP_ID }}
          app_code: ${{ secrets.ENDTEST_APP_CODE }}
          api_request: >-
            https://app.endtest.io/api.php?action=runWeb&appId=${{ secrets.ENDTEST_APP_ID }}&appCode=${{ secrets.ENDTEST_APP_CODE }}&label=Critical&platform=windows&os=windows11&browser=chrome&browserVersion=latest&resolution=1280x1024&geolocation=sanfrancisco&cases=all&notes=
          number_of_loops: 60
          results_format: json-light
          fail_build: true

      - name: Show the execution summary
        if: always()
        env:
          EXECUTION_COUNT: ${{ steps.endtest.outputs.execution_count }}
          TEST_CASES: ${{ steps.endtest.outputs.test_cases }}
          PASSED: ${{ steps.endtest.outputs.passed }}
          FAILED: ${{ steps.endtest.outputs.failed }}
          ERRORS: ${{ steps.endtest.outputs.errors }}
          FAILED_TEST_CASES: ${{ steps.endtest.outputs.failed_test_cases }}
          TEST_EXECUTIONS: ${{ steps.endtest.outputs.test_executions }}
        run: |
          echo "Executions: $EXECUTION_COUNT"
          echo "Test cases: $TEST_CASES"
          echo "Passed assertions: $PASSED"
          echo "Failed assertions: $FAILED"
          echo "Errors: $ERRORS"
          echo "Failed test cases: $FAILED_TEST_CASES"
          echo "$TEST_EXECUTIONS" | jq

• app_id {string}, required: The App ID for your Endtest account, available on the Endtest Settings page.
• app_code {string}, required: The App Code for your Endtest account, available on the Endtest Settings page.
• api_request {string}, required: The complete Endtest API request that starts the test execution or executions.
• number_of_loops {integer}, required: The maximum number of result checks. The action waits 30 seconds before each check.
• results_format {string}, optional: json for detailed logs and media URLs, or json-light for a smaller summary response. The default is json.
• fail_build {boolean}, optional: Set to true to fail the GitHub Actions step when any execution reports failed assertions or errors. The default is false for backward compatibility.

For CI workflows, setting fail_build: true is recommended. The action writes all outputs before failing the step, so later steps using if: always() can still publish or inspect the Endtest results.

The step fails when the aggregated failed value is greater than zero, the aggregated errors value is greater than zero, a test case status is Failed, Error, or Erred, or the Endtest API itself returns Erred..

Leave fail_build set to false for informational or non-blocking test executions.

When Endtest returns multiple hashes, the action sends all hashes in one getResults request and waits until the API returns a result for every hash.

The existing outputs remain convenient for single executions and behave as follows for multiple executions:

• test_suite_name and configuration become JSON arrays.
• test_cases, passed, failed, and errors are totals across all executions.
• start_time is the earliest start time and end_time is the latest end time.
• detailed_logs and screenshots_and_video are flattened JSON arrays.
• results becomes a JSON array of individual Endtest Results page URLs.

Use test_executions when you need the complete per-execution data without aggregation.

• test_suite_name {string}: One test suite name, or a JSON array of names.
• configuration {string}: One configuration, or a JSON array of configurations.
• test_cases {integer}: Total test cases across all executions.
• passed {integer}: Total passed assertions across all executions.
• failed {integer}: Total failed assertions across all executions.
• errors {integer}: Total errors across all executions.
• failed_test_cases {integer}: Total test cases with a Failed, Error, or Erred status.
• start_time {timestamp}: Earliest execution start time.
• end_time {timestamp}: Latest execution end time.
• detailed_logs {JSON string}: Flattened detailed logs. This is an empty array with json-light.
• screenshots_and_video {JSON string}: Flattened screenshot, log, and video URLs. This is an empty array with json-light.
• hash {string}: The original hash response, including comma-separated hashes.
• results {string}: One Results page URL, or a JSON array of URLs.
• execution_count {integer}: Number of completed executions.
• hashes {JSON string}: Array of execution hashes.
• result_urls {JSON string}: Array of Results page URLs.
• test_executions {JSON string}: Complete API response normalized to an array.

The action fails with a clear message when:

• The trigger request does not return a valid hash or comma-separated hash list.
• Endtest returns Erred..
• The results response is not valid JSON.
• Not all execution results are available before number_of_loops is exhausted.
• fail_build is true and Endtest reports failed assertions, errors, or failed test case statuses.