Enumerations

Defines the possible final outcomes of a scan operation. Type: (str, Enum)

Member	Value	Description
`PASSED`	`"passed"`	The scan completed and no policies were violated.
`FLAGGED`	`"flagged"`	A policy with a `FLAG` action was triggered.
`BLOCKED`	`"blocked"`	A policy with a `BLOCK` action was triggered.

Defines the action to be taken when a policy’s conditions are met. Type: (str, Enum)

Member	Value	Description
`BLOCK`	`"block"`	Instructs the system to issue a `BLOCKED` verdict.
`FLAG`	`"flag"`	Instructs the system to issue a `FLAGGED` verdict.
`DISABLED`	`"disabled"`	The policy will be ignored during the scan.

Specifies the context of the text being provided for a scan. Type: (str, Enum)

Member	Value	Description
`PROMPT`	`"PROMPT"`	The provided text is a user-generated prompt.
`RESPONSE`	`"RESPONSE"`	The provided text is an LLM-generated response.

Defines the unique string identifiers for each available security policy. Type: (str, Enum)

Member	Value
`POLICY_VIOLATION`	`"policy_violation_detection"`
`SECRETS_KEYS`	`"secrets_keys_detection"`
`PII_DETECTION`	`"pii_detection"`
`TOXICITY`	`"toxicity_detection"`
`COMPETITOR_MENTION`	`"competitor_mention_detection"`
`BANNED_TOPICS`	`"banned_topics_detection"`
`PROMPT_INJECTION`	`"prompt_injection_detection"`
`UNSAFE_PROMPT`	`"unsafe_prompt_protection"`
`UNSAFE_RESPONSE`	`"unsafe_response_detection"`
`SYSTEM_PROMPT_LEAK`	`"system_prompt_leak_detection"`
`LANGUAGE_VIOLATION`	`language_violation_detection`

Represents the abstract type of a node in an agentic workflow. Type: (str, Enum)

Member	Value	Description
`ORCHESTRATOR`	`"ORCHESTRATOR"`	A node that coordinates other nodes in the workflow.
`AGENT`	`"AGENT"`	A node that performs actions or makes decisions.
`TOOL`	`"TOOL"`	A node that represents a specific capability or function.
`GUARDRAILS`	`"GUARDRAILS"`	A node that enforces safety and security policies.

Represents the specific subtype of a node in an agentic workflow.