v1

latestOpenAPI 3.0.12026-07-226992320.0 KB

QA Rules

Create QA Rule

Creates a new QA rule with the specified configuration. Define the rule's evaluation criteria and expected result. Returns the created QA rule, including its qaRuleId for subsequent get, update, delete, and issue-listing calls.

post/v1/qa-rules

Request body

ruleNamestring required

The name of the QA rule, used to identify it in rule lists and evaluation reports.

ruleDescriptionstring

A brief explanation of what this QA rule evaluates and its intended purpose.

evaluationCriteriastring required

The specific criteria used to evaluate agent responses against this rule.

expectedValuetrue | false

The expected outcome of the evaluation criteria that determines a successful QA result. Set to true if the evaluation criteria returning true should be treated as a Success, or false if the evaluation criteria returning true should be treated as a Fail. Default is true.

agentIdsstring[] required

List of agent Ids on which QA rule would apply

Example request

{
  "ruleName": "Tone & Professionalism Check",
  "ruleDescription": "Ensures agents maintain a professional and empathetic tone throughout the conversation.",
  "evaluationCriteria": "Evaluate whether agent greeted the customer by name, avoided interrupting, and closed with a resolution summary.",
  "expectedValue": true,
  "agentIds": [
    "D5D0p7TUs66TTAEAx"
  ]
}

Response

qaRuleIdstring

The unique identifier of the QA rule.

workspaceIdstring

The unique identifier of the workspace this QA rule belongs to.

ruleNamestring

The name of the QA rule, used to identify it in rule lists and evaluation reports.

ruleDescriptionstring

A brief explanation of what this QA rule evaluates and its intended purpose.

evaluationCriteriastring

The specific criteria or rubric used to evaluate agent responses against this rule.

expectedValueboolean

The expected evaluation result that determines whether the QA rule is satisfied. When the evaluation criteria returns a result matching this expected value, the interaction is marked as a Success; otherwise it is marked as a Fail.

Example response

{
  "qaRuleId": "rule_YYdKgw1masomMfe8",
  "workspaceId": "org_V6eZ2zg8ziDx5pft",
  "ruleName": "Tone & Professionalism Check",
  "ruleDescription": "Ensures agents maintain a professional and empathetic tone throughout the conversation.",
  "agents": [
    {
      "name": "New AI Agent",
      "agentId": "agent_D5D0p7TUs66TTAEAx",
      "status": "Live"
    }
  ],
  "evaluationCriteria": "Evaluate whether agent greeted the customer by name, avoided interrupting, and closed with a resolution summary.",
  "expectedValue": true
}