Generate Evaluator
Generate test scenarios in the background
Authorizations
OAuth access token issued by Cekura for connected apps.
Body
Number of evaluators to generate (1-100)
1 <= x <= 100ID of the agent to generate evaluator for
Alternative to agent_id - the assistant ID to generate evaluators for
List of personality IDs to use for generated evaluators
ID of the test profile to assign to generated evaluators
Type of scenarios to generate: workflow (standard scenarios), red_teaming_voice (red teaming with voice-suitable prompts only), or red_teaming_text (red teaming with all prompts)
workflow- workflowred_teaming_voice- red_teaming_voicered_teaming_text- red_teaming_text
workflow, red_teaming_voice, red_teaming_text Attack category for red teaming scenarios. Required when scenario_type is red_teaming_voice or red_teaming_text. Categories: system_prompt_leak (attempts to extract the system prompt), data_leak (attempts to extract sensitive data), harmful_content (attempts to generate harmful output), biased_output (attempts to generate biased responses), unauthorized_actions (attempts to trigger unauthorized tool calls or actions), off_task (attempts to derail the conversation away from the agent's purpose)
- `` -
system_prompt_leak- system_prompt_leakdata_leak- data_leakharmful_content- harmful_contentbiased_output- biased_outputunauthorized_actions- unauthorized_actionsoff_task- off_task
, system_prompt_leak, data_leak, harmful_content, biased_output, unauthorized_actions, off_task, null Custom expected outcome for the scenario. If not provided, a default is generated based on the attack type.
Additional instructions to guide evaluator generation
Files used only as context for this generation run. Supports PDF, TXT, JSON, CSV, XML, and Markdown; the combined upload size cannot exceed 50 MB.
10Output format for generated workflow evaluators. Use instruction for prompt-based evaluators or conditional_actions for structured condition/action evaluators.
instruction- instructionconditional_actions- conditional_actions
instruction, conditional_actions Phone number ID to use for inbound evaluator
Initial message to start the conversation
1000If true, expected outcomes will be generated for each evaluator
List of tags to apply to all generated scenarios
List of tool IDs to use for generated evaluators
Folder path to assign generated scenarios to (e.g., 'folder_name' or 'parent.child')