NeuralTrust | The leading security platform for generative AI

Overview

The AWS Bedrock Guardrail plugin is a robust content filtering system that leverages Amazon Bedrock’s guardrail capabilities to protect your AI gateway. It provides comprehensive content moderation through Amazon’s pre-trained models and customizable guardrails, ensuring that content passing through your gateway adheres to specified policies and guidelines. The plugin integrates seamlessly with AWS Bedrock’s guardrail system, which offers three primary types of protection:

Topic Policy: Controls allowed and disallowed topics
Content Policy: Filters potentially harmful or inappropriate content
Sensitive Information Policy: Detects and protects against exposure of sensitive data

Each policy type can be configured through AWS Bedrock’s guardrail configuration, making this plugin highly flexible and adaptable to various use cases and security requirements.

Features

Topic-based Filtering:

Block specific topics or subject matters
Customizable topic policies
Fine-grained control over allowed content

Content Moderation:

Detection of harmful content
Customizable content policies
Multiple filter types for different content categories

PII Protection:

Detection of sensitive information
Protection against data leakage
Customizable PII entity types

Action Configuration:

Customizable block messages
Detailed error reporting
Comprehensive logging

Authentication

The AWS Bedrock Guardrail plugin supports two authentication methods for connecting to AWS services. Proper authentication configuration is essential for the plugin to access AWS Bedrock guardrail services.

Authentication Methods

1. AWS Access Key/Secret Key Authentication

This is the traditional method using AWS access keys. Best suited for development environments or when using dedicated service accounts.

{
    "settings": {
        "guardrail_id": "<your-guardrail-id>",
        "version": "1",
        "credentials": {
            "aws_access_key": "AKIA...",
            "aws_secret_key": "your-secret-key",
            "aws_region": "us-east-1",
            "aws_session_token": "optional-session-token"
        },
        "actions": {
            "message": "%s"
        }
    }
}

2. IAM Role-Based Authentication (Recommended)

This method uses AWS STS (Security Token Service) to assume an IAM role. This is the recommended approach for production environments as it provides better security through temporary credentials.

{
    "settings": {
        "guardrail_id": "<your-guardrail-id>",
        "version": "1",
        "credentials": {
            "use_role": true,
            "role_arn": "arn:aws:iam::123456789012:role/TrustGateBedrockRole",
            "session_name": "trustgate-bedrock-session",
            "aws_region": "us-east-1"
        },
        "actions": {
            "message": "%s"
        }
    }
}

Authentication Configuration Parameters

Credentials Configuration

Parameter	Description	Required	Authentication Method
aws_access_key	AWS access key ID	Yes*	Access Key/Secret
aws_secret_key	AWS secret access key	Yes*	Access Key/Secret
aws_region	AWS region for Bedrock service	Yes	Both
aws_session_token	Optional AWS session token for temporary creds	No	Access Key/Secret
use_role	Enable IAM role-based authentication	No	Role-based
role_arn	ARN of the IAM role to assume	Yes**	Role-based
session_name	Name for the assumed role session	No	Role-based

*Required when use_role is false or not specified
**Required when use_role is true

Authentication Best Practices

Security Recommendations

Use IAM Roles in Production: Role-based authentication is more secure as it uses temporary credentials and follows AWS security best practices.

Principle of Least Privilege: Ensure the IAM user or role has only the minimum permissions required:

{
    "Version": "2012-10-17",
    "Statement": [
        {
            "Effect": "Allow",
            "Action": [
                "bedrock:ApplyGuardrail"
            ],
            "Resource": [
                "arn:aws:bedrock:*:*:guardrail/*"
            ]
        }
    ]
}

Region Selection: Choose the AWS region closest to your deployment for optimal performance and compliance.
Credential Management:
- Never hardcode credentials in configuration files
- Use environment variables or secure credential management systems
- Rotate access keys regularly when using access key authentication

Environment-Specific Configurations

Development Environment:

{
    "credentials": {
        "aws_access_key": "${AWS_ACCESS_KEY_ID}",
        "aws_secret_key": "${AWS_SECRET_ACCESS_KEY}",
        "aws_region": "us-east-1"
    }
}

Production Environment:

{
    "credentials": {
        "use_role": true,
        "role_arn": "arn:aws:iam::${AWS_ACCOUNT_ID}:role/TrustGateBedrockRole",
        "session_name": "trustgate-bedrock-prod",
        "aws_region": "us-east-1"
    }
}

Configuration Examples

Basic Configuration

The basic configuration provides essential content filtering with a specified guardrail:

{
    "settings": {
        "guardrail_id": "<your-guardrail-id>",
        "version": "1",
        "actions": {
            "message": "%s"
        }
    }
}

Key components of the basic configuration:

Configuration Parameters

Parameter	Description	Required
guardrail_id	AWS Bedrock Guardrail identifier	Yes
version	Guardrail version (defaults to “1”)	No
actions.message	Custom message template for blocked content	No

Security-Focused Configuration

For environments requiring strict security controls:

{
    "settings": {
        "guardrail_id": "<your-guardrail-id>",
        "version": "2",
        "actions": {
            "message": "Security violation detected: %s. This incident has been logged."
        }
    }
}

Example Use Cases

1. Content Moderation Gateway

Create a gateway that moderates content using AWS Bedrock’s guardrails:

curl -X POST "http://localhost:8080/api/v1/gateways" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "Bedrock Guardrail Gateway",
    "subdomain": "bedrock-guard",
    "required_plugins": [
        {
            "name": "bedrock_guardrail",
            "enabled": true,
            "stage": "pre_request",
            "priority": 1,
            "settings": {
                "guardrail_id": "<your-guardrail-id>",
                "version": "1",
                "actions": {
                    "message": "%s"
                }
            }
        }
    ]
}'

Test with safe content:

curl -X POST "http://localhost:8081/post" \
    -H "Host: bedrock-guard.example.com" \
    -H "X-TG-API-Key: ${API_KEY}" \
    -H "Content-Type: text/plain" \
    -d "The latest deployment includes performance improvements and bug fixes."

# Expected Response: 200 OK

Test with policy violation:

curl -X POST "http://localhost:8081/post" \
    -H "Host: bedrock-guard.example.com" \
    -H "X-TG-API-Key: ${API_KEY}" \
    -H "Content-Type: text/plain" \
    -d "I dont like obama"

# Expected Response: 403 Forbidden
# Message: "Content blocked: Potentially harmful content detected"

2. Topic Policy Protection

The topic policy feature allows you to control which subjects or topics are allowed or denied in the content. When a topic is blocked, the plugin will return a specific error message indicating which topic violated the policy. Example configuration focusing on topic policies:

curl -X POST "http://localhost:8080/api/v1/gateways" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "Topic Protection Gateway",
    "subdomain": "topic-protection",
    "required_plugins": [
        {
            "name": "bedrock_guardrail",
            "enabled": true,
            "stage": "pre_request",
            "priority": 1,
            "settings": {
                "guardrail_id": "<your-guardrail-id>",
                "version": "1",
                "actions": {
                    "message": "Topic policy violation"
                }
            }
        }
    ]
}'

3. Sensitive Information Protection

The plugin can detect and block content containing sensitive information such as PII (Personally Identifiable Information):

curl -X POST "http://localhost:8080/api/v1/gateways" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "PII Protection Gateway",
    "subdomain": "pii-protection",
    "required_plugins": [
        {
            "name": "bedrock_guardrail",
            "enabled": true,
            "stage": "pre_request",
            "priority": 1,
            "settings": {
                "guardrail_id": "<your-guardrail-id>",
                "version": "1",
                "actions": {
                    "message": "Sensitive information detected"
                }
            }
        }
    ]
}'

Best Practices

1. Guardrail Configuration

• Use appropriate guardrail IDs for your use case • Test guardrails thoroughly before deployment • Keep guardrail versions consistent across environments

2. Error Handling

• Implement descriptive error messages • Log policy violations appropriately • Review and update policies regularly

3. Performance Optimization

• Use appropriate AWS region for reduced latency

Performance Considerations

The AWS Bedrock Guardrail plugin is designed for optimal performance while providing robust content filtering capabilities. The plugin maintains a single AWS client instance per plugin instance, reducing the overhead of creating new connections for each request. The implementation includes efficient error handling and logging mechanisms to ensure reliable operation in production environments. Key performance features include:

Efficient AWS client management
Optimized request handling
Comprehensive logging for debugging
Minimal memory footprint
Fast response times for content evaluation

The plugin’s architecture supports concurrent processing with low latency, making it suitable for high-throughput environments. The integration with AWS Bedrock’s guardrail system ensures that content evaluation is performed quickly and efficiently, with minimal impact on overall system performance.

Getting Started

Core Concepts

Traffic Management

Actions API

Non-REST Connectivity

Rate Limiting & Request Control

Content Security

Application Security

Server Security

Data masking

Extending Functionality

Observability & Monitoring

Benchmark

API Reference

AWS Bedrock Guardrail

Overview

Features

Authentication

Authentication Methods

1. AWS Access Key/Secret Key Authentication

2. IAM Role-Based Authentication (Recommended)

Authentication Configuration Parameters

Credentials Configuration

Authentication Best Practices

Security Recommendations

Environment-Specific Configurations

Configuration Examples

Basic Configuration

Configuration Parameters

Security-Focused Configuration

Example Use Cases

1. Content Moderation Gateway

2. Topic Policy Protection

3. Sensitive Information Protection

Best Practices

1. Guardrail Configuration

2. Error Handling

3. Performance Optimization

Performance Considerations

Getting Started

Core Concepts

Traffic Management

Actions API

Non-REST Connectivity

Rate Limiting & Request Control

Content Security

Application Security

Server Security

Data masking

Extending Functionality

Observability & Monitoring

Benchmark

API Reference

​Overview

​Features

​Authentication

​Authentication Methods

​1. AWS Access Key/Secret Key Authentication

​2. IAM Role-Based Authentication (Recommended)

​Authentication Configuration Parameters

​Credentials Configuration

​Authentication Best Practices

​Security Recommendations

​Environment-Specific Configurations

​Configuration Examples

​Basic Configuration

​Configuration Parameters

​Security-Focused Configuration

​Example Use Cases

​1. Content Moderation Gateway

​2. Topic Policy Protection

​3. Sensitive Information Protection

​Best Practices

​1. Guardrail Configuration

​2. Error Handling

​3. Performance Optimization

​Performance Considerations

Overview

Features

Authentication

Authentication Methods

1. AWS Access Key/Secret Key Authentication

2. IAM Role-Based Authentication (Recommended)

Authentication Configuration Parameters

Credentials Configuration

Authentication Best Practices

Security Recommendations

Environment-Specific Configurations

Configuration Examples

Basic Configuration

Configuration Parameters

Security-Focused Configuration

Example Use Cases

1. Content Moderation Gateway

2. Topic Policy Protection

3. Sensitive Information Protection

Best Practices

1. Guardrail Configuration

2. Error Handling

3. Performance Optimization

Performance Considerations