Skip to main content

​
check_policy_violation()

Scans text against a custom list of keywords or rules. This policy is applicable to both prompts and responses. Signature: 
def check_policy_violation(self, text: str, interaction_type: InteractionType, action: Action, rules: List[str], **kwargs) -> ApiResult
Parameters:
  • text (str): The input text to scan.
  • interaction_type (InteractionType): The context of the text (PROMPT or RESPONSE).
  • action (Action): The action to take if a rule is matched (BLOCK or FLAG).
  • rules (List[str]): A list of forbidden keywords or phrases to detect.
  • *kwargs: Optional platform overrides (asset_id, session_id, save).

​
check_secrets_keys()

Scans text for hardcoded secrets and keys. This policy is applicable only to responses. Signature: 
def check_secrets_keys(self, text: str, action: Action, patterns: Optional[List[Tuple[str, str]]] = None, **kwargs) -> ApiResult
Parameters:
  • text (str): The input text to scan.
  • action (Action): The action to take (BLOCK or FLAG).
  • patterns (Optional[List[Tuple[str, str]]]): A list of (name, regex_pattern) tuples to whitelist specific patterns, preventing them from being flagged as secrets.
  • *kwargs: Optional platform overrides (asset_id, session_id, save).

​
check_pii()

Scans text for Personally Identifiable Information. This policy is applicable to both prompts and responses. Signature: 
def check_pii(self, text: str, interaction_type: InteractionType, action: Action, **kwargs) -> ApiResult
Parameters:
  • text (str): The input text to scan.
  • interaction_type (InteractionType): The context of the text (PROMPT or RESPONSE).
  • action (Action): The action to take (BLOCK or FLAG).
  • *kwargs: Optional platform overrides (asset_id, session_id, save).

​
check_toxicity()

Scans text for toxic content. This policy is applicable to both prompts and responses. Signature: 
def check_toxicity(self, text: str, interaction_type: InteractionType, action: Action, **kwargs) -> ApiResult
Parameters:
  • text (str): The input text to scan.
  • interaction_type (InteractionType): The context of the text (PROMPT or RESPONSE).
  • action (Action): The action to take (BLOCK or FLAG).
  • *kwargs: Optional platform overrides (asset_id, session_id, save).

​
check_competitor_mention()

Scans text for mentions of specific competitor names. This policy is applicable to both prompts and responses. Signature: 
def check_competitor_mention(self, text: str, interaction_type: InteractionType, action: Action, competitors: List[str], **kwargs) -> ApiResult
Parameters:
  • text (str): The input text to scan.
  • interaction_type (InteractionType): The context of the text (PROMPT or RESPONSE).
  • action (Action): The action to take (BLOCK or FLAG).
  • competitors (List[str]): A list of competitor names to detect.
  • *kwargs: Optional platform overrides (asset_id, session_id, save).

​
check_banned_topics()

Scans text for forbidden topics. This policy is applicable only to prompts. Signature: 
def check_banned_topics(self, prompt: str, action: Action, topics: List[str], **kwargs) -> ApiResult
Parameters:
  • prompt (str): The input prompt to scan.
  • action (Action): The action to take (BLOCK or FLAG).
  • topics (List[str]): A list of forbidden topics.
  • *kwargs: Optional platform overrides (asset_id, session_id, save).

​
check_prompt_injection()

Scans text for prompt injection attacks. This policy is applicable only to prompts. Signature: 
def check_prompt_injection(self, prompt: str, action: Action, **kwargs) -> ApiResult
Parameters:
  • prompt (str): The input prompt to scan.
  • action (Action): The action to take (BLOCK or FLAG).
  • *kwargs: Optional platform overrides (asset_id, session_id, save).

​
check_unsafe_prompt()

Scans text for requests for harmful or unethical content. This policy is applicable only to prompts. Signature: 
def check_unsafe_prompt(self, prompt: str, action: Action, **kwargs) -> ApiResult
Parameters:
  • prompt (str): The input prompt to scan.
  • action (Action): The action to take (BLOCK or FLAG).
  • *kwargs: Optional platform overrides (asset_id, session_id, save).

​
check_unsafe_response()

Scans text to ensure it does not contain harmful or unethical content. This policy is applicable only to responses. Signature: 
def check_unsafe_response(self, text: str, action: Action, **kwargs) -> ApiResult
Parameters:
  • text (str): The input response to scan.
  • action (Action): The action to take (BLOCK or FLAG).
  • *kwargs: Optional platform overrides (asset_id, session_id, save).

​
check_system_prompt_leak()

Scans text for content matching a confidential system prompt. This policy is applicable only to responses. Signature: 
def check_system_prompt_leak(self, text: str, action: Action, system_prompt: str, **kwargs) -> ApiResult
Parameters:
  • text (str): The input response to scan.
  • action (Action): The action to take (BLOCK or FLAG).
  • system_prompt (str): The system prompt string to check for leakage against.
  • *kwargs: Optional platform overrides (asset_id, session_id, save).