V2.09
Powered by

# Getting Started

This guide walks you through running your first AI security test with Teev.

# Prerequisites

Before starting, ensure you have:

  • ✅ Completed onboarding and account setup
  • ✅ Added your AI application to Teev
  • ✅ Selected relevant test scenarios
  • ✅ Verified connection to your AI system

# Running Your First Test

# Step 1: Navigate to Testing

  1. Log into your Teev dashboard
  2. Select your AI application from the dropdown
  3. Click "Run Test" or navigate to Testing > New Test

# Step 2: Configure Test Parameters

Select Test Scenarios

  • Choose from available test groups based on your AI application type
  • Start with basic scenarios like "Security Vulnerabilities" for your first test
  • You can always run additional scenarios later

Set Test Scope

  • Quick Test: ~50 probes, completes in 5-10 minutes
  • Standard Test: ~200 probes, completes in 15-30 minutes
  • Comprehensive Test: ~500+ probes, completes in 45-90 minutes

Configure Options

  • Test intensity (conservative, standard, aggressive)
  • Timeout settings
  • Notification preferences

# Step 3: Launch the Test

  1. Review your configuration
  2. Click "Start Test"
  3. Monitor progress in real-time through the dashboard
  4. Receive notifications when testing completes

# Understanding Results

# Test Results Dashboard

Your results will include:

Overall Security Score

  • High-level assessment of your AI application's security posture
  • Score from 0-100 with higher scores indicating better security

Vulnerability Breakdown

  • Critical, High, Medium, and Low severity issues
  • Detailed descriptions of each finding
  • Specific examples of failed test cases

Risk Categories

  • Issues grouped by type (jailbreaks, data leakage, etc.)
  • Impact assessment for each category
  • Recommended prioritization

# Taking Action

Review Critical Issues First

  1. Start with Critical and High severity vulnerabilities
  2. Click on individual findings for detailed information
  3. Review the provided remediation guidance
  4. Implement fixes in your AI application

Track Progress

  1. Re-run tests after implementing fixes
  2. Compare results over time
  3. Monitor your security score improvements

# Best Practices

# Regular Testing

  • Set up automated weekly or monthly tests
  • Test after any significant changes to your AI application
  • Monitor for new threats and updated test scenarios

# Comprehensive Coverage

  • Gradually expand testing to include all relevant scenarios
  • Consider your specific use case and risk profile
  • Test different user personas and interaction patterns

# Team Collaboration

  • Share results with development and security teams
  • Use reporting features to demonstrate security improvements
  • Establish processes for addressing findings

# Common Issues

Connection Problems

  • Verify API endpoints and authentication
  • Check firewall and network connectivity
  • Review API rate limits and quotas

Unexpected Results

  • Ensure your AI application is functioning normally
  • Review test scenario configurations
  • Check for recent changes to your system

# Next Steps

After your first successful test:

  1. Explore Advanced Features: Review Tips and Tricks
  2. Set Up Monitoring: Configure automated testing schedules
  3. Integrate Workflows: Connect Teev with your CI/CD pipeline
  4. Expand Testing: Add more comprehensive test scenarios

Need help? Check our FAQ or contact support.