strands.types.guardrails

Guardrail-related type definitions for the SDK.

These types are modeled after the Bedrock API.

• Bedrock docs: https://docs.aws.amazon.com/bedrock/latest/APIReference/API_Types_Amazon_Bedrock_Runtime.html

GuardrailConfig

class GuardrailConfig(TypedDict)

Defined in: src/strands/types/guardrails.py:13

Configuration for content filtering guardrails.

Attributes:

• guardrailIdentifier - Unique identifier for the guardrail.
• guardrailVersion - Version of the guardrail to apply.
• streamProcessingMode - Processing mode.
• trace - The trace behavior for the guardrail.

Topic

class Topic(TypedDict)

Defined in: src/strands/types/guardrails.py:29

Information about a topic guardrail.

Attributes:

• action - The action the guardrail should take when it intervenes on a topic.
• name - The name for the guardrail.
• type - The type behavior that the guardrail should perform when the model detects the topic.

TopicPolicy

class TopicPolicy(TypedDict)

Defined in: src/strands/types/guardrails.py:43

A behavior assessment of a topic policy.

Attributes:

• topics - The topics in the assessment.

ContentFilter

class ContentFilter(TypedDict)

Defined in: src/strands/types/guardrails.py:53

The content filter for a guardrail.

Attributes:

• action - Action to take when content is detected.
• confidence - Confidence level of the detection.
• type - The type of content to filter.

ContentPolicy

class ContentPolicy(TypedDict)

Defined in: src/strands/types/guardrails.py:67

An assessment of a content policy for a guardrail.

Attributes:

• filters - List of content filters to apply.

CustomWord

class CustomWord(TypedDict)

Defined in: src/strands/types/guardrails.py:77

Definition of a custom word to be filtered.

Attributes:

• action - Action to take when the word is detected.
• match - The word or phrase to match.

ManagedWord

class ManagedWord(TypedDict)

Defined in: src/strands/types/guardrails.py:89

Definition of a managed word to be filtered.

Attributes:

• action - Action to take when the word is detected.
• match - The word or phrase to match.
• type - Type of the word.

WordPolicy

class WordPolicy(TypedDict)

Defined in: src/strands/types/guardrails.py:103

The word policy assessment.

Attributes:

• customWords - List of custom words to filter.
• managedWordLists - List of managed word lists to filter.

PIIEntity

class PIIEntity(TypedDict)

Defined in: src/strands/types/guardrails.py:115

Definition of a Personally Identifiable Information (PII) entity to be filtered.

Attributes:

• action - Action to take when PII is detected.
• match - The specific PII instance to match.
• type - The type of PII to detect.

Regex

class Regex(TypedDict)

Defined in: src/strands/types/guardrails.py:161

Definition of a custom regex pattern for filtering sensitive information.

Attributes:

• action - Action to take when the pattern is matched.
• match - The regex filter match.
• name - Name of the regex pattern for identification.
• regex - The regex query.

SensitiveInformationPolicy

class SensitiveInformationPolicy(TypedDict)

Defined in: src/strands/types/guardrails.py:177

Policy defining sensitive information filtering rules.

Attributes:

• piiEntities - List of Personally Identifiable Information (PII) entities to detect and handle.
• regexes - The regex queries in the assessment.

ContextualGroundingFilter

class ContextualGroundingFilter(TypedDict)

Defined in: src/strands/types/guardrails.py:189

Filter for ensuring responses are grounded in provided context.

Attributes:

• action - Action to take when the threshold is not met.
• score - The score generated by contextual grounding filter (range [0, 1]).
• threshold - Threshold used by contextual grounding filter to determine whether the content is grounded or not.
• type - The contextual grounding filter type.

ContextualGroundingPolicy

class ContextualGroundingPolicy(TypedDict)

Defined in: src/strands/types/guardrails.py:205

The policy assessment details for the guardrails contextual grounding filter.

Attributes:

• filters - The filter details for the guardrails contextual grounding filter.

GuardrailAssessment

class GuardrailAssessment(TypedDict)

Defined in: src/strands/types/guardrails.py:215

A behavior assessment of the guardrail policies used in a call to the Converse API.

Attributes:

• contentPolicy - The content policy.
• contextualGroundingPolicy - The contextual grounding policy used for the guardrail assessment.
• sensitiveInformationPolicy - The sensitive information policy.
• topicPolicy - The topic policy.
• wordPolicy - The word policy.

GuardrailTrace

class GuardrailTrace(TypedDict)

Defined in: src/strands/types/guardrails.py:233

Trace information from guardrail processing.

Attributes:

• inputAssessment - Assessment of input content against guardrail policies, keyed by input identifier.
• modelOutput - The original output from the model before guardrail processing.
• outputAssessments - Assessments of output content against guardrail policies, keyed by output identifier.

Trace

class Trace(TypedDict)

Defined in: src/strands/types/guardrails.py:247

A Top level guardrail trace object.

Attributes:

• guardrail - Trace information from guardrail processing.