Looking for automatic AI-powered PR reviews? See GitHub App instead.
1
Configure Secrets
Add the following secrets to your GitHub repository:
QATECH_API_TOKEN
- Your QA.tech API tokenQATECH_PROJECT_ID
- Your QA.tech project ID
2
Create Workflow File
Create This runs all tests in your project whenever code is pushed to
.github/workflows/qatech.yml
in your repository:main
. The blocking
option waits for test results before completing the workflow.3
Trigger Your First Run
Push to your main branch or open a PR (depending on your workflow configuration). QA.tech will execute your tests and report results directly in the GitHub Actions UI.
Basic Usage
The simplest way to run tests is without any additional configuration:Testing Pull Requests
Run tests automatically when PRs are opened or updated:- Catch issues before merge
- Validate changes against existing tests
- Gate PRs based on test results
Testing Preview Deployments
Test preview environments with dynamic URLs:- Testing Vercel/Netlify preview deployments
- Multi-environment testing (frontend + backend)
- PR-specific URL testing
Scheduled Testing
Run tests on a schedule (e.g., nightly regression suite):- Comprehensive nightly regression testing
- Production smoke tests
- Load testing during off-peak hours
Direct API Integration
If you prefer direct API calls instead of the GitHub Action, you can use the REST API.Basic Request
With Environment Overrides
Exploratory Testing with @qatech
Private Beta - This feature is currently in private beta. Request access by emailing daniel@qa.tech
@qatech
in a PR comment. The AI agent analyzes your changes, discovers functionality gaps, creates tests, and provides detailed feedback.
This is different from automatic PR reviews. Exploratory testing is on-demand and focuses on discovering issues through interactive testing, while automatic reviews run on every PR.
@qatech
anywhere in a PR comment:
- Analyze changes - Examines PR diff and crawls the preview deployment
- Check coverage - Identifies which existing tests apply
- Create tests - Generates new tests for untested functionality
- Run tests - Executes relevant tests against the PR preview
- Post review - Writes a detailed GitHub review with findings
Use Exploratory Testing | Use Automatic PR Reviews |
---|---|
Deep investigation of complex changes | Every PR as standard gating |
Interactive testing with custom instructions | Hands-off automated coverage |
Specific areas need thorough validation | General functional regression |
Debugging test failures | Consistent baseline testing |
- “Test the entire checkout flow with different payment methods”
- “Explore what happens when users upload large files”
- “Check if the new search handles edge cases”
- Organization name
- GitHub repository
- Use case (what you want to test)
Configuration Reference
GitHub Action Parameter Reference
Inputs
Input | Description | Required | Default |
---|---|---|---|
project_id | Your QA.tech project ID | Yes | - |
api_token | QA.tech API token | Yes | - |
test_plan_short_id | Test plan short ID to run | No | All tests |
blocking | Wait for test results before completing | No | false |
applications_config | JSON with application environment overrides | No | - |
api_url | Custom API URL | No | https://app.qa.tech |
Outputs
Output | Description |
---|---|
runId | The ID of the created test run |
runShortId | Short ID for the test run |
success | Boolean indicating if run was successful |
run_result | Test execution result (PASSED , FAILED , SKIPPED ) - only when blocking: true |
run_status | Final run status (COMPLETED , ERROR , etc.) - only when blocking: true |
Related Features
- Automatic PR reviews: GitHub App runs tests on every PR
- Direct API: Start Run API Reference for custom integrations