azure-ai-evaluation 1.0.0b2__py3-none-any.whl → 1.13.3__py3-none-any.whl

azure/ai/evaluation/red_team/_red_team_result.py

@@ -0,0 +1,661 @@
+# ---------------------------------------------------------
+# Copyright (c) Microsoft Corporation. All rights reserved.
+# ---------------------------------------------------------
+from typing import Any, Dict, List, Optional, TypedDict
+import json
+from azure.ai.evaluation._common._experimental import experimental
+
+
+# NOTE: The RedTeam* TypedDict classes below are being gradually deprecated in favor of
+# the standard EvalRun* models from azure.ai.evaluation._common.onedp.models._models.
+# New code should use EvalRunOutputItem instead of RedTeamOutputItem, etc.
+# These TypedDicts are maintained for backward compatibility during the transition.
+
+
+@experimental
+class RiskCategorySummary(TypedDict):
+    """Summary of attack success rates across risk categories.
+
+    :param overall_asr: Overall attack success rate across all risk categories
+    :type overall_asr: float
+    :param overall_total: Total number of attacks across all risk categories
+    :type overall_total: int
+    :param overall_successful_attacks: Number of successful attacks across all risk categories
+    :type overall_successful_attacks: int
+    :param hate_unfairness_asr: Attack success rate for hate/unfairness risk category
+    :type hate_unfairness_asr: float
+    :param hate_unfairness_total: Total number of attacks for hate/unfairness risk category
+    :type hate_unfairness_total: int
+    :param hate_unfairness_successful_attacks: Number of successful attacks for hate/unfairness risk category
+    :type hate_unfairness_successful_attacks: int
+    :param violence_asr: Attack success rate for violence risk category
+    :type violence_asr: float
+    :param violence_total: Total number of attacks for violence risk category
+    :type violence_total: int
+    :param violence_successful_attacks: Number of successful attacks for violence risk category
+    :type violence_successful_attacks: int
+    :param sexual_asr: Attack success rate for sexual risk category
+    :type sexual_asr: float
+    :param sexual_total: Total number of attacks for sexual risk category
+    :type sexual_total: int
+    :param sexual_successful_attacks: Number of successful attacks for sexual risk category
+    :type sexual_successful_attacks: int
+    :param self_harm_asr: Attack success rate for self-harm risk category
+    :type self_harm_asr: float
+    :param self_harm_total: Total number of attacks for self-harm risk category
+    :type self_harm_total: int
+    :param self_harm_successful_attacks: Number of successful attacks for self-harm risk category
+    :type self_harm_successful_attacks: int
+    """
+
+    overall_asr: float
+    overall_total: int
+    overall_successful_attacks: int
+    hate_unfairness_asr: float
+    hate_unfairness_total: int
+    hate_unfairness_successful_attacks: int
+    violence_asr: float
+    violence_total: int
+    violence_successful_attacks: int
+    sexual_asr: float
+    sexual_total: int
+    sexual_successful_attacks: int
+    self_harm_asr: float
+    self_harm_total: int
+    self_harm_successful_attacks: int
+
+
+@experimental
+class AttackTechniqueSummary(TypedDict):
+    """Summary of attack success rates across complexity levels.
+
+    :param overall_asr: Overall attack success rate across all complexity levels
+    :type overall_asr: float
+    :param overall_total: Total number of attacks across all complexity levels
+    :type overall_total: int
+    :param overall_successful_attacks: Number of successful attacks across all complexity levels
+    :type overall_successful_attacks: int
+    :param baseline_asr: Attack success rate for baseline complexity level
+    :type baseline_asr: float
+    :param baseline_total: Total number of attacks for baseline complexity level
+    :type baseline_total: int
+    :param baseline_successful_attacks: Number of successful attacks for baseline complexity level
+    :type baseline_successful_attacks: int
+    :param easy_complexity_asr: Attack success rate for easy complexity level
+    :type easy_complexity_asr: float
+    :param easy_complexity_total: Total number of attacks for easy complexity level
+    :type easy_complexity_total: int
+    :param easy_complexity_successful_attacks: Number of successful attacks for easy complexity level
+    :type easy_complexity_successful_attacks: int
+    :param moderate_complexity_asr: Attack success rate for moderate complexity level
+    :type moderate_complexity_asr: float
+    :param moderate_complexity_total: Total number of attacks for moderate complexity level
+    :type moderate_complexity_total: int
+    :param moderate_complexity_successful_attacks: Number of successful attacks for moderate complexity level
+    :type moderate_complexity_successful_attacks: int
+    :param difficult_complexity_asr: Attack success rate for difficult complexity level
+    :type difficult_complexity_asr: float
+    :param difficult_complexity_total: Total number of attacks for difficult complexity level
+    :type difficult_complexity_total: int
+    :param difficult_complexity_successful_attacks: Number of successful attacks for difficult complexity level
+    :type difficult_complexity_successful_attacks: int
+    """
+
+    overall_asr: float
+    overall_total: int
+    overall_successful_attacks: int
+    baseline_asr: float
+    baseline_total: int
+    baseline_successful_attacks: int
+    easy_complexity_asr: float
+    easy_complexity_total: int
+    easy_complexity_successful_attacks: int
+    moderate_complexity_asr: float
+    moderate_complexity_total: int
+    moderate_complexity_successful_attacks: int
+    difficult_complexity_asr: float
+    difficult_complexity_total: int
+    difficult_complexity_successful_attacks: int
+
+
+@experimental
+class JointRiskAttackSummaryItem(TypedDict):
+    """Summary of attack success rates for a specific risk category across complexity levels.
+
+    :param risk_category: The risk category being summarized
+    :type risk_category: str
+    :param baseline_asr: Attack success rate for baseline complexity level
+    :type baseline_asr: float
+    :param easy_complexity_asr: Attack success rate for easy complexity level
+    :type easy_complexity_asr: float
+    :param moderate_complexity_asr: Attack success rate for moderate complexity level
+    :type moderate_complexity_asr: float
+    :param difficult_complexity_asr: Attack success rate for difficult complexity level
+    :type difficult_complexity_asr: float
+    """
+
+    risk_category: str
+    baseline_asr: float
+    easy_complexity_asr: float
+    moderate_complexity_asr: float
+    difficult_complexity_asr: float
+
+
+@experimental
+class RedTeamingScorecard(TypedDict):
+    """TypedDict representation of a Red Team Agent scorecard with the updated structure.
+
+    :param risk_category_summary: Overall metrics by risk category
+    :type risk_category_summary: List[RiskCategorySummary]
+    :param attack_technique_summary: Overall metrics by attack technique complexity
+    :type attack_technique_summary: List[AttackTechniqueSummary]
+    :param joint_risk_attack_summary: Detailed metrics by risk category and complexity level
+    :type joint_risk_attack_summary: List[JointRiskAttackSummaryItem]
+    :param detailed_joint_risk_attack_asr: Detailed ASR information broken down by complexity level, risk category, and converter
+    :type detailed_joint_risk_attack_asr: Dict[str, Dict[str, Dict[str, float]]]
+    """
+
+    risk_category_summary: List[RiskCategorySummary]
+    attack_technique_summary: List[AttackTechniqueSummary]
+    joint_risk_attack_summary: List[JointRiskAttackSummaryItem]
+    detailed_joint_risk_attack_asr: Dict[str, Dict[str, Dict[str, float]]]
+
+
+@experimental
+class AttackObjectiveSource(TypedDict):
+    """Information about how attack objectives were generated.
+
+    :param application_scenario: The application scenario used for generating attack objectives
+    :type application_scenario: str
+    :param risk_categories: List of risk categories targeted by the attack objectives
+    :type risk_categories: List[str]
+    :param custom_attack_seed_prompts: Custom prompts used to seed attack objective generation
+    :type custom_attack_seed_prompts: str
+    :param policy_document: Policy document used for generating attack objectives
+    :type policy_document: str
+    """
+
+    application_scenario: str
+    risk_categories: List[str]
+    custom_attack_seed_prompts: str
+    policy_document: str
+
+
+@experimental
+class RedTeamingParameters(TypedDict):
+    """TypedDict representation of parameters used in a red team evaluation with the updated structure.
+
+    :param attack_objective_generated_from: Information about the source of attack objectives
+    :type attack_objective_generated_from: AttackObjectiveSource
+    :param attack_complexity: List of complexity levels used in the attacks
+    :type attack_complexity: List[str]
+    :param techniques_used: Dictionary mapping complexity levels to lists of converter techniques used
+    :type techniques_used: Dict[str, List[str]]
+    :param attack_success_thresholds: Thresholds used for determining attack success per risk category
+    :type attack_success_thresholds: Dict[str, Any]
+    """
+
+    attack_objective_generated_from: AttackObjectiveSource
+    attack_complexity: List[str]
+    techniques_used: Dict[str, List[str]]
+    attack_success_thresholds: Dict[str, Dict[str, int]]
+
+
+@experimental
+class RiskAssessment(TypedDict):
+    """Assessment of content safety risks in a conversation.
+
+    :param violence: Assessment of violence-related content
+    :type violence: Dict[str, str]
+    :param sexual: Assessment of sexual content
+    :type sexual: Dict[str, str]
+    :param self_harm: Assessment of self-harm content
+    :type self_harm: Dict[str, str]
+    :param hate_unfair: Assessment of hate or unfairness content
+    :type hate_unfair: Dict[str, str]
+    """
+
+    violence: Dict[str, str]
+    sexual: Dict[str, str]
+    self_harm: Dict[str, str]
+    hate_unfair: Dict[str, str]
+
+
+@experimental
+class AttackDetails(TypedDict, total=False):
+    """TypedDict representation of a simulated conversation in a red team evaluation.
+
+    :param attack_success: Whether the attack was successful
+    :type attack_success: Optional[bool]
+    :param attack_technique: The attack technique used in the conversation
+    :type attack_technique: str
+    :param attack_complexity: The complexity level of the attack
+    :type attack_complexity: str
+    :param risk_category: The risk category targeted by the attack
+    :type risk_category: str
+    :param conversation: List of messages exchanged in the conversation
+    :type conversation: List[Dict[str, str]]
+    :param risk_assessment: Dictionary containing content safety assessment for the conversation
+    :type risk_assessment: Optional[RiskAssessment]
+    :param attack_success_threshold: The threshold value used to determine attack success
+    :type attack_success_threshold: Optional[int]
+    :param risk_sub_type: Optional risk sub-category/sub-type for the attack
+    :type risk_sub_type: Optional[str]
+    """
+
+    attack_success: Optional[bool]
+    attack_technique: str
+    attack_complexity: str
+    risk_category: str
+    conversation: List[Dict[str, str]]
+    risk_assessment: Optional[RiskAssessment]
+    attack_success_threshold: Optional[int]
+    risk_sub_type: Optional[str]
+
+
+@experimental
+class RedTeamOutputResultProperties(TypedDict, total=False):
+    """Additional metadata captured for each evaluation result."""
+
+    attack_success: Optional[bool]
+    attack_success_threshold: Optional[int]
+    attack_technique: str
+    attack_complexity: str
+    risk_category: str
+    risk_assessment: Optional[Dict[str, Any]]
+    reason: Optional[str]
+    severity_label: Optional[str]
+    metadata: Optional[Dict[str, Any]]
+
+
+@experimental
+class EvaluationRunOutputItemMessage(TypedDict, total=False):
+    """Representation of a single message within an evaluation sample."""
+
+    role: str
+    content: Any
+    name: Optional[str]
+    tool_calls: Optional[List[Dict[str, Any]]]
+
+
+@experimental
+class RedTeamRunOutputItemResult(TypedDict, total=False):
+    """Flattened evaluation result for a single risk category.
+
+    :param passed: Indicates whether the attack was unsuccessful (True) or successful (False).
+                   In the context of attack success rate (ASR), passed=True means the system
+                   successfully defended against the attack, while passed=False means the attack
+                   succeeded in compromising the system.
+    :type passed: Optional[bool]
+    :param label: String label "pass" or "fail" that aligns with the passed field
+    :type label: Optional[str]
+    """
+
+    # Should extend EvaluationRunOutputItemResult
+
+    object: str
+    type: str
+    name: str
+    passed: Optional[bool]
+    label: Optional[str]
+    score: Optional[float]
+    metric: Optional[str]
+    threshold: Optional[float]
+    reason: Optional[str]
+    properties: RedTeamOutputResultProperties
+
+
+@experimental
+class RedTeamDatasourceItem(TypedDict, total=False):
+    """Metadata about the datasource item that produced this conversation."""
+
+    id: Optional[str]
+    input_data: Dict[str, Any]
+
+
+@experimental
+class RedTeamRunOutputItemSample(TypedDict, total=False):
+    """Sample payload containing the red team conversation.
+
+    :param error: Error information from either the evaluation step or while calling the target system.
+                  Contains details about any failures that occurred during the attack simulation or
+                  evaluation process.
+    :type error: Optional[Dict[str, Any]]
+    """
+
+    # Should extend EvaluationRunOutputItemSample
+
+    object: str
+    input: List[EvaluationRunOutputItemMessage]
+    output: List[EvaluationRunOutputItemMessage]
+    finish_reason: Optional[str]
+    model: Optional[str]
+    error: Optional[Dict[str, Any]]
+    usage: Optional[Dict[str, Any]]
+    seed: Optional[int]
+    temperature: Optional[float]
+    top_p: Optional[float]
+    max_completion_tokens: Optional[float]
+    metadata: Optional[Dict[str, Any]]
+
+
+@experimental
+class RedTeamOutputItem(TypedDict, total=False):
+    """Structured representation of a conversation and its evaluation artifacts.
+
+    DEPRECATED: This TypedDict duplicates the EvalRunOutputItem model from
+    azure.ai.evaluation._common.onedp.models._models. New code should use
+    EvalRunOutputItem directly instead of this TypedDict wrapper.
+    """
+
+    object: str
+    id: str
+    created_time: int
+    status: str
+    datasource_item_id: Optional[str]
+    datasource_item: Optional[RedTeamDatasourceItem]
+    sample: RedTeamRunOutputItemSample
+    results: List[RedTeamRunOutputItemResult]
+
+
+@experimental
+class ScanResult(TypedDict):
+    """TypedDict representation of a Red Team Agent evaluation result with the updated structure.
+
+    :param scorecard: Scorecard containing summary and detailed ASR information
+    :type scorecard: RedTeamingScorecard
+    :param parameters: Parameters containing metadata about the evaluation run
+    :type parameters: RedTeamingParameters
+    :param attack_details: List of AttackDetails objects representing the conversations in the evaluation
+    :type attack_details: List[AttackDetails]
+    :param AOAI_Compatible_Row_Results: List of evaluation results for each risk category
+    :type AOAI_Compatible_Row_Results: Optional[List[RedTeamRunOutputItemResult]]
+    :param AOAI_Compatible_Summary: The evaluation run metadata in eval.run format
+    :type AOAI_Compatible_Summary: Optional[RedTeamRun]
+    :param studio_url: Optional URL for the studio
+    :type studio_url: Optional[str]
+    """
+
+    scorecard: RedTeamingScorecard
+    parameters: RedTeamingParameters
+    attack_details: List[AttackDetails]
+    AOAI_Compatible_Row_Results: Optional[List[RedTeamRunOutputItemResult]]
+    AOAI_Compatible_Summary: Optional["RedTeamRun"]
+    studio_url: Optional[str]
+
+
+@experimental
+class ResultCount(TypedDict):
+    """Count of evaluation results by status.
+
+    :param total: Total number of evaluation results
+    :type total: int
+    :param passed: Number of evaluation results where the attack was unsuccessful (system defended successfully).
+                   In the context of attack success rate (ASR), this represents attacks that failed.
+    :type passed: int
+    :param failed: Number of evaluation results where the attack was successful (system was compromised).
+                   In the context of attack success rate (ASR), this represents successful attacks.
+    :type failed: int
+    :param errored: Number of evaluation results that failed to process due to errors.
+                    Calculated from rows where the sample's error property is set, indicating failures
+                    during the attack simulation or evaluation step.
+    :type errored: int
+    """
+
+    total: int
+    passed: int
+    failed: int
+    errored: int
+
+
+@experimental
+class PerTestingCriteriaResult(TypedDict, total=False):
+    """Result count for a specific testing criteria.
+
+    :param testing_criteria: The name of the testing criteria (e.g., risk category)
+    :type testing_criteria: str
+    :param attack_strategy: The attack strategy used (optional, for attack strategy summaries)
+    :type attack_strategy: Optional[str]
+    :param passed: Number of results where the attack was unsuccessful for this criteria.
+                   In the context of attack success rate (ASR), this represents attacks that failed
+                   to compromise the system for this specific testing criteria.
+    :type passed: int
+    :param failed: Number of results where the attack was successful for this criteria.
+                   In the context of attack success rate (ASR), this represents successful attacks
+                   for this specific testing criteria.
+    :type failed: int
+    """
+
+    testing_criteria: str
+    attack_strategy: Optional[str]
+    passed: int
+    failed: int
+
+
+@experimental
+class DataSourceItemGenerationParams(TypedDict, total=False):
+    """Parameters for data source item generation.
+
+    :param type: Type of data source generation (e.g., "red_team")
+    :type type: str
+    :param attack_strategies: List of attack strategies used
+    :type attack_strategies: List[str]
+    :param num_turns: Number of turns in the conversation
+    :type num_turns: int
+    """
+
+    type: str
+    attack_strategies: List[str]
+    num_turns: int
+
+
+@experimental
+class DataSource(TypedDict, total=False):
+    """Data source information for the red team evaluation.
+
+    :param type: Type of data source (e.g., "azure_ai_red_team")
+    :type type: str
+    :param target: Target configuration for the data source
+    :type target: Dict[str, Any]
+    :param item_generation_params: Parameters used for generating data items
+    :type item_generation_params: DataSourceItemGenerationParams
+    """
+
+    type: str
+    target: Dict[str, Any]
+    item_generation_params: DataSourceItemGenerationParams
+
+
+@experimental
+class OutputItemsList(TypedDict):
+    """Wrapper for list of output items.
+
+    :param object: Object type identifier (always "list")
+    :type object: str
+    :param data: List of red team output items
+    :type data: List[RedTeamOutputItem]
+    """
+
+    object: str
+    data: List[RedTeamOutputItem]
+
+
+@experimental
+class RedTeamRun(TypedDict, total=False):
+    """TypedDict representation of a Red Team evaluation run in eval.run format.
+
+    :param object: Object type identifier (always "eval.run")
+    :type object: str
+    :param id: Unique identifier for the run
+    :type id: str
+    :param eval_id: Identifier for the evaluation experiment
+    :type eval_id: str
+    :param created_at: Timestamp when the run was created (Unix epoch seconds)
+    :type created_at: int
+    :param status: Status of the run (e.g., "completed", "failed", "in_progress")
+    :type status: str
+    :param name: Display name for the run
+    :type name: str
+    :param report_url: URL to view the run report in Azure AI Studio
+    :type report_url: Optional[str]
+    :param data_source: Information about the data source used for the evaluation
+    :type data_source: DataSource
+    :param metadata: Additional metadata for the run
+    :type metadata: Dict[str, Any]
+    :param result_counts: Aggregated counts of evaluation results
+    :type result_counts: ResultCount
+    :param per_model_usage: Usage statistics per model (if applicable)
+    :type per_model_usage: List[Any]
+    :param per_testing_criteria_results: Results aggregated by testing criteria
+    :type per_testing_criteria_results: List[PerTestingCriteriaResult]
+    :param output_items: Wrapped list of output items from the evaluation
+    :type output_items: OutputItemsList
+    :param conversations: Optional list of attack details/conversations
+    :type conversations: List[AttackDetails]
+    """
+
+    object: str
+    id: str
+    eval_id: str
+    created_at: int
+    status: str
+    name: str
+    report_url: Optional[str]
+    data_source: DataSource
+    metadata: Dict[str, Any]
+    result_counts: ResultCount
+    per_model_usage: List[Any]
+    per_testing_criteria_results: List[PerTestingCriteriaResult]
+    output_items: OutputItemsList
+    conversations: List[AttackDetails]
+
+
+@experimental
+class RedTeamResult:
+    def __init__(self, scan_result: Optional[ScanResult] = None, attack_details: Optional[List[AttackDetails]] = None):
+        self.scan_result = scan_result
+        self.attack_details = attack_details
+
+    def to_json(self) -> str:
+        """
+        Converts a RedTeamResult object to a JSON-serializable dictionary.
+
+        :returns: A string containing the RedTeamResult in JSON format.
+        :rtype: str
+        """
+        return json.dumps(self.scan_result) if self.scan_result else ""
+
+    def to_scorecard(self) -> Optional[RedTeamingScorecard]:
+        """Extracts the scorecard from a RedTeamResult object."""
+        return self.scan_result.get("scorecard", None) if self.scan_result else None
+
+    def to_eval_qr_json_lines(self) -> str:
+        """Converts conversations in messages format to query-response format suitable for evaluation.
+
+        The output format follows the JSONL pattern with each line containing:
+
+        .. code-block:: javascript
+
+            {
+                "query": "user message content",
+                "response": "assistant message content",
+                "risk_category": "risk category",
+                "attack_strategy": "strategy name",
+                "attack_complexity": "complexity level",
+                "attack_success": "true|false", // (if available from evaluation)
+                "category": "risk category", // (if available from evaluation)
+                "severity_level": "low|medium|high", // (if available from evaluation)
+                "threshold": "threshold value" // (if available from evaluation)
+            }
+
+        :returns: A list of strings containing query-response pairs in JSONL format.
+        :rtype: List[str]
+        """
+        if not self.attack_details:
+            return ""
+
+        result_lines = []
+
+        for conversation in self.attack_details:
+            messages = conversation.get("conversation", [])
+            attack_technique = conversation.get("attack_technique", "")
+            attack_complexity = conversation.get("attack_complexity", "")
+            risk_category = conversation.get("risk_category", "")
+            attack_success = conversation.get("attack_success")
+            risk_assessment = conversation.get("risk_assessment", {})
+
+            for i in range(0, len(messages) - 1, 2):
+                if i + 1 < len(messages):
+                    user_msg = messages[i]
+                    assistant_msg = messages[i + 1]
+
+                    if user_msg.get("role") == "user" and assistant_msg.get("role") == "assistant":
+                        # Create the base query-response pair
+                        qr_pair = {
+                            "query": user_msg.get("content", ""),
+                            "response": assistant_msg.get("content", ""),
+                            "risk_category": risk_category,
+                            "attack_strategy": attack_technique,
+                            "attack_complexity": attack_complexity,
+                        }
+
+                        # Add evaluation data if available
+                        if attack_success is not None:
+                            qr_pair["attack_success"] = str(attack_success).lower()
+
+                        # Add risk assessment data if available
+                        for category, assessment in risk_assessment.items() if risk_assessment else {}:
+                            if assessment and assessment.get("severity_label", None):
+                                qr_pair["category"] = category
+                                qr_pair["severity_level"] = assessment.get("severity_label", "").lower()
+                                # Add threshold if available in the future
+
+                        result_lines.append(json.dumps(qr_pair))
+
+        return result_lines
+
+    def attack_simulation(self) -> str:
+        """
+        Returns the attack simulation data in a human-readable format.
+        :returns: A string containing the attack simulation data in a human-readable format.
+        :rtype: str
+        """
+        if not self.attack_details:
+            return ""
+
+        result_lines = []
+
+        for conversation in self.attack_details:
+            messages = conversation.get("conversation", [])
+            attack_technique = conversation.get("attack_technique", "")
+            attack_complexity = conversation.get("attack_complexity", "")
+            risk_category = conversation.get("risk_category", "")
+            attack_success = conversation.get("attack_success")
+            risk_assessment = conversation.get("risk_assessment", {})
+
+            result_lines.append(f"Attack Technique: {attack_technique}")
+            result_lines.append(f"Attack Complexity: {attack_complexity}")
+            result_lines.append(f"Risk Category: {risk_category}")
+            result_lines.append("")
+
+            for i in range(0, len(messages) - 1, 2):
+                if i + 1 < len(messages):
+                    user_msg = messages[i]
+                    assistant_msg = messages[i + 1]
+
+                    if user_msg.get("role") == "user" and assistant_msg.get("role") == "assistant":
+                        result_lines.append(f"User: {user_msg.get('content', '')}")
+                        result_lines.append(f"Assistant: {assistant_msg.get('content', '')}")
+                        result_lines.append("")
+
+            if attack_success is not None:
+                result_lines.append(f"Attack Success: {'Successful' if attack_success else 'Failed'}")
+                result_lines.append("")
+
+            for category, assessment in risk_assessment.items() if risk_assessment else {}:
+                if assessment and assessment.get("severity_label", None):
+                    result_lines.append(f"Category: {category}")
+                    result_lines.append(f"Severity Level: {assessment.get('severity_label', '')}")
+                    result_lines.append("")
+
+        return "\n".join(result_lines)