Skip to main content
The bugster run command executes your test specifications using an AI agent in automated browser instances. It provides comprehensive test execution with real browser automation and intelligent result analysis.

​
Basic Usage

bugster run

​
Command Overview

Executes test specifications using AI-powered browser automation to validate your application’s functionality across different scenarios and user flows.

​
When to Use

  • Before deploying to validate functionality
  • In CI/CD pipelines for automated testing
  • During development to catch regressions
  • To verify fixes after bug reports

​
Key Features

  • AI agent executes tests in real browser instances
  • Supports parallel execution for faster results
  • Automatic video recording for failed tests
  • Real-time streaming to dashboard
  • Quick ad-hoc testing with text prompts
Requires your application to be running and accessible on the configured URL before executing tests.

​
Command Syntax

bugster run [path] [options]

​
How It Works

The run command:
  1. Test Discovery: Scans for test specifications in .bugster/tests/ directory
  2. Browser Initialization: Launches browser instances (visible or headless)
  3. AI Execution: AI agent interprets and executes test steps
  4. Result Analysis: Captures screenshots, videos, and detailed logs
  5. Reporting: Streams results to dashboard and generates reports

​
Execution Options

​
--headless
boolean
Run tests in headless mode without visible browser UI. Ideal for CI/CD environments.
​
--parallel
integer
default:"3"
Maximum number of concurrent tests to run simultaneously. Higher values speed up execution but use more resources.
​
--max-concurrent
integer
Alias for --parallel. Specifies the maximum number of tests to run concurrently.
​
--limit
integer
Maximum number of tests to execute. Useful for quick validation or debugging specific test counts.
​
--browser
string
default:"chromium"
Browser engine to use for test execution. Supports chromium, firefox, and webkit.
​
--prompt
string
Run an instant test from a text description (max 1000 characters). Creates and executes a test without needing YAML files.
​
--quick
string
Alias for --prompt. Run a quick test from natural language description.

​
Filtering Options

​
--only-affected
boolean
Run only tests for files/directories that have been modified. Perfect for PR workflows and focused testing.
​
--base-url
string
Override the base URL from configuration. Useful for testing against different environments.

​
Output Options

​
--output
string
Save detailed test results to a JSON file. Essential for CI/CD integration and test reporting.
​
--stream-results
boolean
default:"true"
Stream real-time results to the Bugster dashboard for live monitoring.
​
--no-stream-results
boolean
Disable streaming to dashboard. Results will only be available locally.
​
--verbose
boolean
Enable detailed logging output for debugging test execution issues.
​
--silent
boolean
Run in silent mode with minimal output. Reduces noise in CI/CD logs.

​
Examples

​
Complete Test Suite

bugster run
Runs all available tests with default settings and visible browser.

​
Headless CI/CD Execution

bugster run --headless --stream-results
Perfect for automated testing in continuous integration pipelines.

​
Parallel Execution

bugster run --parallel 8 --headless
Runs up to 8 tests concurrently for maximum speed.

​
Target Specific Directory

bugster run auth/ --verbose
Runs only authentication tests with detailed debugging output.

​
Quick Ad-hoc Testing

bugster run --prompt "Test user login with valid credentials and verify dashboard loads"
Creates and runs an instant test from natural language description.

​
Environment Testing

bugster run --base-url https://staging.myapp.com --limit 5
Tests against staging environment with a limited number of tests.

​
Quick Test Prompts

  • Authentication
  • E-commerce
  • Form Validation
  • Mobile Testing
bugster run --prompt "Test user login with valid credentials and verify dashboard loads"
Complete login flow validation.

​
Common Usage Patterns

Development Testing

bugster run
Run all tests with visible browser for development.

CI/CD Pipeline

bugster run --headless --output results.json
Automated testing with result export.

Quick Validation

bugster run --limit 5 --parallel 3
Fast subset testing for quick feedback.

Debugging

bugster run auth/ --verbose
Detailed logging for specific test categories.

​
Test Results

After execution, you’ll receive:
  • Pass/Fail Status: Clear indication of test outcomes
  • Screenshots: Visual evidence of test execution
  • Video Recordings: Automatic recording for failed tests
  • Detailed Logs: Step-by-step execution information
  • Performance Metrics: Timing and resource usage data

​
Best Practices

Use --only-affected in PR workflows to test only changed functionality. Combine --parallel with --headless for maximum execution speed in CI/CD environments.
Quick prompt tests are limited to 1000 characters and run with maximum concurrency of 1. They’re perfect for exploratory testing and quick validation scenarios.
⌘I