> ## Documentation Index > Fetch the complete documentation index at: https://docs.qualifire.ai/llms.txt > Use this file to discover all available pages before exploring further. # Evaluate > Evaluates given input, output, or messages using Qualifire's detectors. Supports checks for hallucinations, grounding, PII, prompt injections, content moderation, policy assertions, topic scoping, and tool use quality. ## OpenAPI ````yaml POST /v1/evaluation/evaluate openapi: 3.0.3 info: title: Qualifire AI API description: API for Qualifire AI services, including evaluation and prompt management. version: 1.0.0 servers: - url: https://api.qualifire.ai/api security: - ApiKeyAuth: [] paths: /v1/evaluation/evaluate: post: tags: - Evaluation summary: Evaluate a prompt using Qualifire's detectors description: >- Evaluates given input, output, or messages using Qualifire's detectors. Supports checks for hallucinations, grounding, PII, prompt injections, content moderation, policy assertions, topic scoping, and tool use quality. requestBody: description: Evaluation request content: application/json: schema: $ref: '#/components/schemas/models.ApiEvaluationEvaluateRequest' required: true responses: '200': description: OK content: application/json: schema: $ref: '#/components/schemas/models.EvaluationResponse' '401': description: Unauthorized content: application/json: schema: $ref: '#/components/schemas/models.Error' '422': description: Validation Error content: application/json: schema: $ref: '#/components/schemas/models.Error' components: schemas: models.ApiEvaluationEvaluateRequest: type: object description: At least one of `input`, `output`, or `messages` must be provided. properties: input: type: string description: The input text (e.g., user prompt) output: type: string description: The output text (e.g., LLM response) messages: type: array description: Conversation messages for multi-turn evaluation items: $ref: '#/components/schemas/models.LlmMessage' available_tools: type: array description: >- List of tools available to the LLM. Required when tool_use_quality_check is enabled. items: $ref: '#/components/schemas/models.LlmToolDefinition' hallucinations_check: type: boolean description: Check for factual inaccuracies or hallucinations hallucinations_mode: type: string description: Quality/speed tradeoff for hallucination checks enum: - speed - balanced - quality grounding_check: type: boolean description: Check if the output is grounded in the provided input/context grounding_mode: type: string description: Quality/speed tradeoff for grounding checks enum: - speed - balanced - quality grounding_multi_turn_mode: type: boolean description: Use full conversation context for grounding checks prompt_injections: type: boolean description: Check for prompt injection attempts pii_check: type: boolean description: Check for personally identifiable information content_moderation_check: type: boolean description: >- Check for harmful content (dangerous content, harassment, hate speech, sexual content) tool_use_quality_check: type: boolean description: >- Check the quality of tool selection and usage. Requires messages and available_tools. tuq_mode: type: string description: Quality/speed tradeoff for tool use quality checks enum: - speed - balanced - quality assertions: type: array description: >- Custom policy assertions to evaluate against (e.g., "don't give medical advice") items: type: string assertions_mode: type: string description: Quality/speed tradeoff for assertion checks enum: - speed - balanced - quality policy_target: type: string description: Whether policy assertions apply to input, output, or both enum: - input - output - both policy_multi_turn_mode: type: boolean description: Use full conversation context for policy checks policy_include_tools: type: boolean description: >- Include tool definitions and tool calls in policy assertion context. When true, assertions can reference available_tools and tool_calls from messages. allowed_topics: type: array description: List of allowed topics for topic scoping checks items: type: string topic_scoping_mode: type: string description: Quality/speed tradeoff for topic scoping checks enum: - speed - balanced - quality topic_scoping_target: type: string description: Whether topic scoping applies to input, output, or both enum: - input - output - both topic_scoping_multi_turn_mode: type: boolean description: Use full conversation context for topic scoping checks metadata: type: object additionalProperties: type: string description: >- Optional key-value pairs (string values only) to attach to the evaluation invocation models.EvaluationResponse: type: object required: - status - score - evaluationResults properties: status: type: string description: Overall evaluation status enum: - success - warning - failed score: type: number nullable: true description: Overall evaluation score (0-100) evaluationResults: type: array items: $ref: '#/components/schemas/models.EvaluationResult' models.Error: type: object required: - error properties: error: type: string description: A human-readable error message example: Invalid API key models.LlmMessage: type: object required: - role properties: role: type: string description: The role of the message sender enum: - system - user - assistant - tool - developer content: type: string nullable: true description: The text content of the message tool_calls: type: array description: Tool calls made by the assistant items: $ref: '#/components/schemas/models.LlmToolCall' models.LlmToolDefinition: type: object required: - name properties: name: type: string description: Name of the tool description: type: string description: Description of what the tool does parameters: type: object additionalProperties: true description: JSON Schema describing the tool's parameters models.EvaluationResult: type: object required: - type - results properties: type: type: string description: >- The evaluation type (e.g., hallucinations, grounding, policy, pii, prompt_injections, safety, tuq, topic_scoping) results: type: array items: $ref: '#/components/schemas/models.DetectionResult' models.LlmToolCall: type: object required: - function properties: id: type: string description: Unique identifier for the tool call type: type: string description: The type of tool call (e.g., "function") function: type: object required: - name - arguments properties: name: type: string description: Name of the tool being called arguments: type: object additionalProperties: true description: Arguments passed to the tool (key-value pairs) models.DetectionResult: type: object required: - name - score - label - confidence_score - reason - flagged properties: name: type: string score: type: number label: type: string confidence_score: type: number reason: type: string quote: type: string flagged: type: boolean description: Whether this result was flagged as a violation data: type: string description: Additional context data (e.g., the assertion text for policy checks) nullable: true securitySchemes: ApiKeyAuth: type: apiKey in: header name: X-Qualifire-API-Key ````