Everything You Need to Secure Your AI
From detection to protection - a complete security suite for LLM applications
Advanced Detectors
150+ specialized detectors for PII, secrets, API keys, credit cards, SSNs, and sensitive data patterns.
Policy Engine
YAML/JSON policy-as-code with 50+ compliance templates (HIPAA, GDPR, PCI-DSS, SOC2). Auto-applied global defaults.
Secure Vault
AES-256 encrypted tokenization and redaction. Securely store and rehydrate sensitive data with format-preserving masking.
AI Gateway
OpenAI-compatible proxy with response filtering, smart routing, rate limiting, and cost tracking. Supports streaming.
Entity Management
Define custom data types, sensitivity levels, and handling rules. Priority-based conflict resolution system.
AI Tool Interceptor
Inspect and control AI agent tool calls. Block dangerous operations, sanitize arguments, and enforce security policies.
Filesystem Guard
Secure file access controls with path validation, content scanning, and automatic sensitive data masking.
Real-time Alerts
Multi-channel notifications (Email, Slack, Webhook). Configurable thresholds, rate limiting, and severity-based routing.
Analytics & Metrics
Prometheus metrics, OpenTelemetry tracing, and comprehensive dashboards. Track usage, costs, and security events.
DAG Workflows
Build complex security pipelines with directed acyclic graphs. Parallel execution, caching, and conditional logic.
Multi-tenant
Complete tenant isolation with role-based access control (RBAC). Admin console for managing organizations.
Easy Integration
Node.js & Python SDKs. REST API, OpenAI-compatible endpoints, and webhook support. Docker & Kubernetes ready.
Flexible Deployment
Self-host on your infrastructure or use our managed cloud service.
Production Ready
Battle-tested in production. Handles millions of requests per day.
Lightning Fast
Sub-10ms latency with intelligent caching and optimization.
Secure Your AI in Minutes
Four simple steps to production-ready AI security
Install SDK
Add GuardrailX to your project with npm or pip
npm install @guardrailx/sdk-nodeInitialize Client
Configure your API key and endpoint
import { GuardrailX } from '@guardrailx/sdk-node';
const client = new GuardrailX({
apiKey: process.env.GUARDRAILX_API_KEY
});Protect Your AI
Wrap your LLM calls with security checks
const result = await client.guard({
input: userMessage,
policy: 'production-policy',
model: 'gpt-4'
});Deploy Securely
Automatic detection, redaction, and compliance
// ✅ PII detected and redacted
// ✅ Policy rules enforced
// ✅ Audit logs created
// ✅ Metrics trackedStart in under 5 minutes • Full documentation included
Built for Every Industry
Pre-configured compliance templates and industry-specific security policies
Enterprise SaaS
Protect customer data in AI chatbots and copilots. Prevent data leaks and ensure compliance.
- Multi-tenant isolation
- RBAC & SSO
- Audit logs
Healthcare
HIPAA-compliant AI with automatic PHI detection and redaction. Secure patient data.
- PHI detection
- HIPAA policies
- Encryption at rest
Financial Services
PCI-DSS compliance for payment data. Detect credit cards, bank accounts, and financial PII.
- PCI-DSS rules
- Fraud detection
- Transaction monitoring
Legal & Compliance
Protect confidential case files and client information with attorney-client privilege controls.
- Document scanning
- Privilege logs
- Retention policies
Education
FERPA-compliant student data protection. Safe AI tutors and educational assistants.
- FERPA policies
- Age verification
- Parent controls
Government
FedRAMP-ready security controls. Protect sensitive government data and citizen information.
- FedRAMP baseline
- NIST controls
- IL4/IL5 ready
Works With Your Stack
Seamless integration with popular LLM providers, languages, and deployment platforms
LLM Providers
OpenAI
GPT-4, GPT-3.5, Embeddings
Anthropic
Claude 3, Claude 2
Cohere
Command, Embed
Hugging Face
Custom models
Azure OpenAI
Enterprise GPT models
AWS Bedrock
Foundation models
Google Vertex AI
PaLM, Gemini
Local Models
Self-hosted LLMs
SDKs & Deployment
Node.js SDK
TypeScript support included
Python SDK
Async/await support
REST API
OpenAPI spec
Docker
Ready-to-deploy images
Kubernetes
Helm charts available
Trusted by Security Teams Worldwide
Trusted by companies worldwide to protect their AI applications with GuardrailX
Sarah Chen
CTO, HealthTech AI
TeleMed Solutions
“GuardrailX helped us achieve HIPAA compliance in weeks instead of months. The PHI detection is incredibly accurate, and the policy engine is flexible enough for complex healthcare workflows.”
Marcus Johnson
VP of Engineering
PaySecure Financial
“We process millions of transactions daily. GuardrailX handles it effortlessly with <5ms latency. The audit logs and compliance reports saved us during our SOC 2 audit.”
Emily Rodriguez
Lead Security Engineer
CloudBot AI
“The multi-tenant isolation is rock solid. We can customize security policies per customer without code changes. The Node.js SDK integration was seamless.”
David Kim
Chief AI Officer
LegalAI Partners
“Attorney-client privilege is non-negotiable for us. GuardrailX detects and protects privileged communications automatically. The workflow engine is a game-changer.”
Priya Patel
Director of Compliance
EduTech Academy
“FERPA compliance was a nightmare before GuardrailX. Now we have automated student data protection across all our AI features. The alerting system gives us peace of mind.”
Alex Turner
Senior DevOps Engineer
DataStream Technologies
“Self-hosted deployment with Docker was straightforward. The metrics and monitoring integrations are excellent. We scaled to 10M+ daily requests without issues.”
Simple, Transparent Pricing
Start with our Starter plan. Scale to millions of requests with enterprise-grade security.
Starter
Get started with essential features
- 150+ Security Detectors
- Policy Engine
- Basic Vault (Local)
- Up to 1M requests/month
- Community Support
- Unlimited tenants
- Multi-tenant isolation
- Advanced Analytics
- Priority Support
- SLA Guarantee
Professional
For growing teams
- Everything in Starter
- Up to 100M requests/month
- Multi-tenant isolation
- Advanced Vault (AWS/Azure)
- Advanced Analytics & Metrics
- Custom Detectors
- Workflow Automation
- Email Support
- 99.9% Uptime SLA
- SOC 2 Ready
Enterprise
For large-scale deployments
- Everything in Professional
- Unlimited requests
- Custom Integrations
- Dedicated Support Engineer
- 99.99% Uptime SLA
- FedRAMP/HIPAA Ready
- On-premise Deployment
- Custom Training
- White-label Option
- MSA & Custom Terms
All plans include core security features. • Flexible billing options available.