Skip to content

Getting Started

Get from zero to a running QA pipeline in under 5 minutes.

Prerequisites

1. Install

pip install ai-testpilot-x

For the visual Streamlit dashboard:

pip install ai-testpilot-x[ui]

For real browser (Selenium) execution:

pip install ai-testpilot-x[selenium]

2. Initialize your project

Run the setup wizard in your project directory:

cd your-project/
testpilot init

This creates testpilot.yaml with your API key, target URL, and execution mode:

# testpilot.yaml
gemini_api_key: ${GEMINI_API_KEY}   # reads from env var
execution_mode: MOCK                 # MOCK | LOCAL | GRID
target_url: https://your-app.com

db_url: sqlite:///.testpilot/testpilot.db
chroma_path: .testpilot/chroma_db
output_dir: .testpilot/reports
log_level: WARNING

API Key

Store your key as an environment variable — never hardcode it: 

export GEMINI_API_KEY="AIzaSy..."
Or add it to a .env file (which is gitignored by default).

3. Run your first pipeline

testpilot run --story "User can log in with valid credentials"

You'll see a live progress display as each agent completes:

  Analyzing requirements...   done  2 modules
  Generating test cases...    done  8 test cases
  Verifying coverage...       done  92% coverage
  Executing tests (MOCK)...   done  8 passed · 0 failed
  Analyzing bugs...           done  0 bugs
  Generating report...        done

  Release Decision:  GO          exit 0
  Risk Score:        12 / 100
  Tests:             8 / 8 passed

4. Execution modes

Mode When to use Requirements
MOCK CI/CD, cloud, no real app None — returns simulated results
LOCAL Local dev with app running Chrome + ChromeDriver installed
GRID Parallel real browser testing Selenium Grid endpoint 
# Mock (default for CI)
testpilot run --story "..." --mode MOCK

# Local Selenium
testpilot run --story "..." --mode LOCAL --url http://localhost:3000

# Selenium Grid
testpilot run --story "..." --mode GRID --url https://staging.myapp.com

5. Analyze test cases without executing

testpilot analyze --story "User can reset their password"

Outputs a table of test cases with priority, type, and steps — no execution needed.

6. Analyze a bug

testpilot bugs --log "NoSuchElementException: Unable to locate element: #login-btn at LoginPage.py:42"

Returns root cause, severity, and fix suggestion with RAG correlation against historical bugs.

7. Launch the visual dashboard

testpilot dashboard

Opens the Streamlit dashboard at http://localhost:8501 with all 8 pages: Test Generator, Selenium Generator, API Tester, Bug Analyzer, Reports, Workflow Studio, Knowledge Base.

Next steps