Fast Setup
Connect your API and start scanning in minutes
No complex configuration required. Just provide your API endpoint and we handle the rest.
Comprehensive Testing
50+ vulnerability patterns tested automatically
Our system tests for prompt injection, data leakage, and security bypasses.
Developer Friendly
Easy integration with your existing workflow
REST API, webhooks, and detailed reports for seamless integration.
Step-by-Step Setup
1
Create Your Account
Sign up for a free PromptShield account to get started.
What you get with the free plan:
- 5 scans per month
- Basic vulnerability detection
- PDF reports
2
Connect Your API
Provide your AI endpoint URL and API key. We support all major LLM providers.
Supported Providers:
OpenAIAnthropicGoogle AIAzure OpenAICustom APIs
Example Configuration:
{
"endpoint": "https://api.openai.com/v1/chat/completions",
"apiKey": "sk-...",
"model": "gpt-4",
"headers": {
"Authorization": "Bearer sk-...",
"Content-Type": "application/json"
}
}3
Run Your First Scan
Start scanning your AI application for vulnerabilities. Our system will test 50+ attack patterns.
What gets tested:
Prompt Injection attacks
Data leakage attempts
Security bypasses
Output validation
4
Review Your Results
Get detailed vulnerability reports with severity ratings, root cause analysis, and remediation steps.
Report includes:
- • Security score (A+ to F)
- • Vulnerability count by severity
- • Detailed attack descriptions
- • Step-by-step remediation guides
- • PDF export for sharing
What's Next?
Explore Advanced Features
Learn about conversation testing, API integration, and custom test patterns.
View API Reference