AWS Bedrock Guardrails

Bifrost integrates with Amazon Bedrock Guardrails to provide enterprise-grade content filtering and safety features with deep AWS integration. This page covers the configuration and capabilities of the AWS Bedrock guardrail provider.

Capabilities

Content Filters: Hate speech, insults, sexual content, violence, misconduct
Denied Topics: Block specific topics or categories
Word Filters: Custom profanity and sensitive word blocking
PII Protection: Detect and redact 50+ PII entity types
Contextual Grounding: Verify responses against source documents
Prompt Attack Detection: Identify injection and jailbreak attempts
Image Content Support: Analyze images in addition to text (PNG, JPEG)

Streaming output: When this profile is used in an output or both rule, Bifrost accumulates the stream until the model response is complete, then checks the full response. It does not check individual stream chunks. See Streaming Output Guardrails for details.

Configuration Fields

Field	Type	Required	Default	Description
`auth_type`	enum	No	inferred	`keys`, `api_key`, or `iam_role`.
`access_key`	string	Conditional	-	Required with `auth_type: keys`.
`secret_key`	string	Conditional	-	Required with `auth_type: keys`.
`session_token`	string	No	-	Optional AWS session token with static credentials.
`bedrock_api_key`	string	Conditional	-	Required with `auth_type: api_key`.
`role_arn`	string	No	-	IAM role ARN to assume after resolving the selected credential source.
`external_id`	string	No	-	Optional external ID for the assumed role.
`session_name`	string	No	-	Optional STS session name for the assumed role.
`guardrail_arn`	string	Yes	-	ARN of the Bedrock guardrail
`guardrail_version`	string	Yes	-	Version of the guardrail (e.g., “1”, “DRAFT”)
`region`	string	No when encoded in `guardrail_arn`	-	AWS region. Bifrost derives it from a full guardrail ARN when omitted.
`images_enabled`	boolean	No	`true`	Send PNG and JPEG image blocks to Bedrock Guardrails.

Authentication

Choose one of the following authentication modes.

Static AWS credentials

{
  "access_key": "AKIAXXXXXXXXXXXXXXXXXX",
  "secret_key": "your-secret-access-key",
  "guardrail_arn": "arn:aws:bedrock:us-east-1:123456789:guardrail/abc123",
  "guardrail_version": "1",
  "region": "us-east-1"
}

Bedrock API key

{
  "auth_type": "api_key",
  "bedrock_api_key": "your-bedrock-api-key",
  "guardrail_arn": "arn:aws:bedrock:us-east-1:123456789:guardrail/abc123",
  "guardrail_version": "1"
}

IAM role or default credential chain

Set auth_type to iam_role. Bifrost uses the AWS SDK default credential chain (for example, IRSA, an EC2/ECS role, environment credentials, or a shared credentials file). Set role_arn to assume a different role when needed.

Supported AWS Regions

The following regions support Amazon Bedrock Guardrails. Availability of individual guardrail policies and safeguard tiers can vary by region; refer to the AWS documentation for the current service matrix.

Region Code	Region Name
`us-east-1`	US East (N. Virginia)
`us-east-2`	US East (Ohio)
`us-west-1`	US West (N. California)
`us-west-2`	US West (Oregon)
`ca-central-1`	Canada (Central)
`eu-central-1`	Europe (Frankfurt)
`eu-north-1`	Europe (Stockholm)
`eu-south-1`	Europe (Milan)
`eu-south-2`	Europe (Spain)
`eu-west-1`	Europe (Ireland)
`eu-west-2`	Europe (London)
`eu-west-3`	Europe (Paris)
`ap-east-2`	Asia Pacific (Taipei)
`ap-northeast-1`	Asia Pacific (Tokyo)
`ap-northeast-2`	Asia Pacific (Seoul)
`ap-south-1`	Asia Pacific (Mumbai)
`ap-southeast-1`	Asia Pacific (Singapore)
`ap-southeast-2`	Asia Pacific (Sydney)
`ap-southeast-3`	Asia Pacific (Jakarta)
`ap-southeast-4`	Asia Pacific (Melbourne)
`ap-southeast-5`	Asia Pacific (Malaysia)
`ap-southeast-7`	Asia Pacific (Thailand)
`me-central-1`	Middle East (UAE)
`il-central-1`	Israel (Tel Aviv)
`us-gov-west-1`	AWS GovCloud (US-West)

Supported Content Types

Text content
Images (PNG, JPEG formats)

Usage Metrics Returned

Bedrock guardrails return detailed usage metrics for cost tracking and monitoring:

Metric	Description
`content_policy_units`	Units consumed by content policy evaluation
`contextual_grounding_policy_units`	Units for grounding checks
`sensitive_information_policy_units`	Units for PII detection
`topic_policy_units`	Units for topic filtering
`word_policy_units`	Units for word filtering
`automated_reasoning_policy_units`	Units for reasoning checks
`content_policy_image_units`	Units for image content analysis

Supported PII Types

Personal identifiers (SSN, passport, driver’s license)
Financial information (credit cards, bank accounts)
Contact information (email, phone, address)
Medical information (health records, insurance)
Device identifiers (IP addresses, MAC addresses)

For provider comparison and information on configuring guardrail rules and profiles, see Guardrails.

Getting Started

Moving from OSS

Release Cadence

Migration Guides

Features

Security

AWS Bedrock Guardrails

Capabilities

Configuration Fields

Authentication

Static AWS credentials

Bedrock API key

IAM role or default credential chain

Supported AWS Regions

Supported Content Types

Usage Metrics Returned

Supported PII Types

​Capabilities

​Configuration Fields

​Authentication

​Static AWS credentials

​Bedrock API key

​IAM role or default credential chain

​Supported AWS Regions

​Supported Content Types

​Usage Metrics Returned

​Supported PII Types

Capabilities

Configuration Fields

Authentication

Static AWS credentials

Bedrock API key

IAM role or default credential chain

Supported AWS Regions

Supported Content Types

Usage Metrics Returned

Supported PII Types