Evaluates messages against inline guardrail checks¶

Description¶

Evaluates messages against inline guardrail checks. You specify the check configurations directly in the request, and Amazon Bedrock returns per-check results with severity or confidence scores.

Usage¶

bedrockruntime_invoke_guardrail_checks(messages, checks)

Arguments¶

messages

[required] The messages to evaluate against the specified guardrail checks. Each message includes a role and one or more content blocks.

checks

[required] The inline check configurations that specify which guardrail checks to run against the messages.

Value¶

A list with the following syntax:

list(
  results = list(
    contentFilter = list(
      results = list(
        list(
          category = "VIOLENCE"|"HATE"|"SEXUAL"|"MISCONDUCT"|"INSULTS",
          severityScore = 123.0
        )
      )
    ),
    promptAttack = list(
      results = list(
        list(
          category = "JAILBREAK"|"PROMPT_INJECTION"|"PROMPT_LEAKAGE",
          severityScore = 123.0
        )
      )
    ),
    sensitiveInformation = list(
      results = list(
        list(
          type = "ADDRESS"|"AGE"|"AWS_ACCESS_KEY"|"AWS_SECRET_KEY"|"CA_HEALTH_NUMBER"|"CA_SOCIAL_INSURANCE_NUMBER"|"CREDIT_DEBIT_CARD_CVV"|"CREDIT_DEBIT_CARD_EXPIRY"|"CREDIT_DEBIT_CARD_NUMBER"|"DRIVER_ID"|"EMAIL"|"INTERNATIONAL_BANK_ACCOUNT_NUMBER"|"IP_ADDRESS"|"LICENSE_PLATE"|"MAC_ADDRESS"|"NAME"|"PASSWORD"|"PHONE"|"PIN"|"SWIFT_CODE"|"UK_NATIONAL_HEALTH_SERVICE_NUMBER"|"UK_NATIONAL_INSURANCE_NUMBER"|"UK_UNIQUE_TAXPAYER_REFERENCE_NUMBER"|"URL"|"USERNAME"|"US_BANK_ACCOUNT_NUMBER"|"US_BANK_ROUTING_NUMBER"|"US_INDIVIDUAL_TAX_IDENTIFICATION_NUMBER"|"US_PASSPORT_NUMBER"|"US_SOCIAL_SECURITY_NUMBER"|"VEHICLE_IDENTIFICATION_NUMBER",
          confidenceScore = 123.0,
          beginOffset = 123,
          endOffset = 123,
          messageIndex = 123,
          contentIndex = 123
        )
      ),
      truncated = TRUE|FALSE
    )
  ),
  usage = list(
    contentFilter = list(
      textUnits = 123
    ),
    promptAttack = list(
      textUnits = 123
    ),
    sensitiveInformation = list(
      textUnits = 123
    )
  )
)

Request syntax¶

svc$invoke_guardrail_checks(
  messages = list(
    list(
      role = "user"|"assistant"|"system",
      content = list(
        list(
          text = "string"
        )
      )
    )
  ),
  checks = list(
    contentFilter = list(
      categories = list(
        list(
          category = "VIOLENCE"|"HATE"|"SEXUAL"|"MISCONDUCT"|"INSULTS"
        )
      )
    ),
    promptAttack = list(
      categories = list(
        list(
          category = "JAILBREAK"|"PROMPT_INJECTION"|"PROMPT_LEAKAGE"
        )
      )
    ),
    sensitiveInformation = list(
      entities = list(
        list(
          type = "ADDRESS"|"AGE"|"AWS_ACCESS_KEY"|"AWS_SECRET_KEY"|"CA_HEALTH_NUMBER"|"CA_SOCIAL_INSURANCE_NUMBER"|"CREDIT_DEBIT_CARD_CVV"|"CREDIT_DEBIT_CARD_EXPIRY"|"CREDIT_DEBIT_CARD_NUMBER"|"DRIVER_ID"|"EMAIL"|"INTERNATIONAL_BANK_ACCOUNT_NUMBER"|"IP_ADDRESS"|"LICENSE_PLATE"|"MAC_ADDRESS"|"NAME"|"PASSWORD"|"PHONE"|"PIN"|"SWIFT_CODE"|"UK_NATIONAL_HEALTH_SERVICE_NUMBER"|"UK_NATIONAL_INSURANCE_NUMBER"|"UK_UNIQUE_TAXPAYER_REFERENCE_NUMBER"|"URL"|"USERNAME"|"US_BANK_ACCOUNT_NUMBER"|"US_BANK_ROUTING_NUMBER"|"US_INDIVIDUAL_TAX_IDENTIFICATION_NUMBER"|"US_PASSPORT_NUMBER"|"US_SOCIAL_SECURITY_NUMBER"|"VEHICLE_IDENTIFICATION_NUMBER"
        )
      )
    )
  )
)