pyrit.scenario.airt.ContentHarms

pyrit.scenario.airt.ContentHarms#

class ContentHarms(*, adversarial_chat: PromptChatTarget | None = None, objective_scorer: TrueFalseScorer | None = None, scenario_result_id: str | None = None, objectives_by_harm: Dict[str, Sequence[SeedGroup]] | None = None)[source]#

Content Harms Scenario implementation for PyRIT.

This scenario contains various harm-based checks that you can run to get a quick idea about model behavior with respect to certain harm categories.

__init__(*, adversarial_chat: PromptChatTarget | None = None, objective_scorer: TrueFalseScorer | None = None, scenario_result_id: str | None = None, objectives_by_harm: Dict[str, Sequence[SeedGroup]] | None = None)[source]#

Initialize the Content Harms Scenario.

Parameters:

adversarial_chat (Optional[PromptChatTarget]) – Additionally used for scoring defaults. If not provided, a default OpenAI target will be created using environment variables.
objective_scorer (Optional[TrueFalseScorer]) – Scorer to evaluate attack success. If not provided, creates a default composite scorer using Azure Content Filter and SelfAsk Refusal scorers. seed_dataset_prefix (Optional[str]): Prefix of the dataset to use to retrieve the objectives. This will be used to retrieve the appropriate seed groups from CentralMemory. If not provided, defaults to “content_harm”.
scenario_result_id (Optional[str]) – Optional ID of an existing scenario result to resume.
objectives_by_harm (Optional[Dict[str, Sequence[SeedGroup]]]) – A dictionary mapping harm strategies to their corresponding SeedGroups. If not provided, default seed groups will be loaded from datasets.

Methods

`__init__`(*[, adversarial_chat, ...])	Initialize the Content Harms Scenario.
`default_dataset_config`()	Return the default dataset configuration for this scenario.
`get_default_strategy`()	Get the default strategy used when no strategies are specified.
`get_strategy_class`()	Get the strategy enum class for this scenario.
`initialize_async`(*[, objective_target, ...])	Initialize the scenario by populating self._atomic_attacks and creating the ScenarioResult.
`run_async`()	Execute all atomic attacks in the scenario sequentially.

Attributes

`atomic_attack_count`	Get the number of atomic attacks in this scenario.
`name`	Get the name of the scenario.
`version`

classmethod default_dataset_config() → DatasetConfiguration[source]#

Return the default dataset configuration for this scenario.

classmethod get_default_strategy() → ScenarioStrategy[source]#

Get the default strategy used when no strategies are specified.

classmethod get_strategy_class() → Type[ScenarioStrategy][source]#

Get the strategy enum class for this scenario.