> ## Documentation Index > Fetch the complete documentation index at: https://docs.valiqor.com/llms.txt > Use this file to discover all available pages before exploring further. # Security Models > Data models for security audit results, red team simulations, batches, vulnerabilities, and attack vectors. ### `SecurityAuditResult` Result of a security audit. | Field | Type | Description | | ---------------------- | ---------------------- | ----------------------------------- | | `audit_id` | `str` | Audit identifier. | | `total_items` | `int` | Total conversations analyzed. | | `safe_count` | `int` | Number of safe items. | | `unsafe_count` | `int` | Number of unsafe items. | | `safety_score` | `float` | Overall safety score (0.0–1.0). | | `triggered_categories` | `Dict[str, int]` | Category → count of violations. | | `top_risk_category` | `Optional[str]` | Most frequently triggered category. | | `details` | `List[Dict[str, Any]]` | Per-item audit details. | **Property:** `overall_safety_score` — alias for `safety_score`. **Alias:** `SecurityEvaluationResult = SecurityAuditResult` (backward compatibility). ### `RedTeamResult` Result of a red team simulation. | Field | Type | Description | | ---------------------- | --------------- | ----------------------------------------------- | | `run_id` | `str` | Run identifier. | | `project_id` | `str` | Project ID. | | `name` | `str` | Run name. | | `status` | `str` | Run status. | | `total_attacks` | `int` | Total attacks generated. | | `successful_attacks` | `int` | Attacks that bypassed defenses. | | `success_rate` | `float` | Attack success rate (0.0–1.0). | | `distinct_vulns_count` | `int` | Number of distinct vulnerabilities found. | | `top_vulnerability` | `Optional[str]` | Most exploited vulnerability. | | `target_url` | `Optional[str]` | Target endpoint URL (if URL mode was used). | | `target_prompt` | `Optional[str]` | Target system prompt (if prompt mode was used). | | `started_at` | `Optional[str]` | ISO timestamp. | **Property:** `attack_success_rate` — alias for `success_rate`. ### `SecurityJobHandle` Wraps an async security job with polling and result retrieval. Same interface as [JobHandle](/sdk/models/evaluation#jobhandle). | Property/Method | Return Type | Description | | ------------------------------------------- | ---------------------------------------- | ---------------------------- | | `job_id` | `str` | Job identifier. | | `job_type` | `str` | `"security"` or `"redteam"`. | | `status()` | `SecurityJobStatus` | Current job status. | | `is_running()` | `bool` | Whether still running. | | `is_completed()` | `bool` | Whether completed. | | `cancel()` | `Dict[str, Any]` | Cancel the job. | | `result()` | `SecurityAuditResult` or `RedTeamResult` | Block and return result. | | `wait(poll_interval, timeout, on_progress)` | `SecurityJobStatus` | Poll with progress. | ### `SecurityJobStatus` | Field | Type | Description | | ------------------ | ------- | ------------------------------------------------------------------ | | `job_id` | `str` | Job identifier. | | `job_type` | `str` | `"security"` or `"redteam"`. | | `status` | `str` | `"queued"`, `"running"`, `"completed"`, `"failed"`, `"cancelled"`. | | `progress_percent` | `float` | Progress (0–100). | | `current_item` | `int` | Current item. | | `total_items` | `int` | Total items. | **Properties:** `is_running`, `is_completed`, `is_failed`. ### `AuditBatch` Summary of an audit batch. | Field | Type | Description | | ------------------- | --------------- | ---------------------- | | `id` | `str` | Batch identifier. | | `project_id` | `str` | Project ID. | | `batch_name` | `Optional[str]` | Batch name. | | `status` | `str` | Batch status. | | `total_items` | `int` | Total items. | | `unsafe_items` | `int` | Unsafe items. | | `unsafe_rate` | `float` | Unsafe rate (0.0–1.0). | | `top_risk_category` | `Optional[str]` | Top risk category. | | `created_at` | `Optional[str]` | ISO timestamp. | **Property:** `safety_score -> float` — computed as `1.0 - unsafe_rate`. ### `AuditBatchItem` A single item from an audit batch. | Field | Type | Description | | -------------------------- | --------------- | -------------------------------------------------- | | `id` | `str` | Item identifier. | | `batch_id` | `str` | Parent batch ID. | | `user_input` | `str` | User message. | | `assistant_response` | `str` | Assistant response. | | `is_safe` | `bool` | Safety determination. | | `triggered_categories` | `List[str]` | Triggered category keys. | | `risk_score` | `float` | Risk score (0.0–1.0). | | `explanation` | `Optional[str]` | Safety explanation. | | `input_safety_categories` | `List[str]` | Safety categories triggered by user input. | | `output_safety_categories` | `List[str]` | Safety categories triggered by assistant response. | ### `AuditBatchItemsPage` Paginated `AuditBatchItem` list. Supports `len()` and iteration. ### `RedTeamRun` | Field | Type | Description | | ---------------------- | --------------- | ------------------------- | | `id` | `str` | Run identifier. | | `project_id` | `str` | Project ID. | | `name` | `str` | Run name. | | `status` | `str` | Status. | | `total_attacks` | `int` | Total attacks. | | `successful_attacks` | `int` | Successful attacks. | | `success_rate` | `float` | Success rate. | | `distinct_vulns_count` | `int` | Distinct vulnerabilities. | | `top_vulnerability` | `Optional[str]` | Top vulnerability. | | `target_url` | `Optional[str]` | Target endpoint URL. | | `target_prompt` | `Optional[str]` | Target system prompt. | ### `RedTeamAttack` | Field | Type | Description | | --------------- | --------------- | ------------------------------------- | | `id` | `str` | Attack identifier. | | `run_id` | `str` | Parent run ID. | | `attack_vector` | `str` | Attack vector used. | | `prompt` | `str` | Attack prompt. | | `response` | `str` | Target response. | | `succeeded` | `bool` | Whether the attack bypassed defenses. | | `vulnerability` | `Optional[str]` | Vulnerability exploited. | | `severity` | `Optional[str]` | Severity level. | | `explanation` | `Optional[str]` | Explanation. | ### `RedTeamAttacksPage` Paginated `RedTeamAttack` list. Supports `len()` and iteration. ### `RedTeamComparison` | Field | Type | | ----------------------- | ---------------------- | | `runs` | `List[Dict[str, Any]]` | | `vectors` | `List[Dict[str, Any]]` | | `overall_success_rates` | `Dict[str, float]` | | `baseline_run_id` | `Optional[str]` | ### `VulnerabilityInfo` | Field | Type | | -------------- | --------------- | | `key` | `str` | | `display_name` | `str` | | `description` | `Optional[str]` | | `severity` | `Optional[str]` | ### `AttackVectorInfo` | Field | Type | | ------------- | --------------- | | `key` | `str` | | `name` | `str` | | `description` | `Optional[str]` | ### `ProjectVulnerability` | Field | Type | Default | | -------------- | --------------- | ------- | | `id` | `str` | — | | `project_id` | `str` | — | | `key` | `str` | — | | `display_name` | `str` | — | | `enabled` | `bool` | `True` | | `threshold` | `float` | `0.5` | | `severity` | `Optional[str]` | `None` | ### `ProjectAttackVector` | Field | Type | Default | | ------------ | ------- | ------- | | `id` | `str` | — | | `project_id` | `str` | — | | `key` | `str` | — | | `name` | `str` | — | | `enabled` | `bool` | `True` | | `weight` | `float` | `1.0` | *** ## Related * [SecurityClient](/sdk/security-client) — Methods returning these models * [Evaluation Models](/sdk/models/evaluation) — Eval data models * [Failure Analysis Models](/sdk/models/failure-analysis) — FA data models