Skip to main content

AIFeatureRun

A dispatched AI feature run. A partial unique index on (featureId, scopeType, scopeId) filtered to non-terminal statuses means only one in-flight run per feature+connector can exist at a time; a second attempt returns 409.

idinteger
featureIdstring
Example: refine_mcp_descriptions
scopeTypeAIFeatureRunScopeType (string)

Resource type an AI feature run applies to. Only connector is supported today.

Possible values: [connector]

scopeIdinteger

The connector ID this run applies to.

jobIdinteger
requestedByinteger

ID of the user who triggered the run.

organizationIdinteger
statusAIFeatureRunStatus (string)

Possible values: [pending, running, retrying, completed, failed, cancelled]

cleanupResources object[]
  • Array [
  • typestring
    Constant value: mcp_configuration
    slugstring
  • ]
  • parsedResultobject | nullnullable

    The worker's structured output, validated against the feature's result schema, present once the run completes.

    estimatedCostUsdnumber | nullnullable
    createdAtstring<date-time>
    updatedAtstring<date-time>
    AIFeatureRun
    {
    "id": 0,
    "featureId": "refine_mcp_descriptions",
    "scopeType": "connector",
    "scopeId": 0,
    "jobId": 0,
    "requestedBy": 0,
    "organizationId": 0,
    "status": "pending",
    "cleanupResources": [
    {
    "type": "mcp_configuration",
    "slug": "string"
    }
    ],
    "parsedResult": "Unknown Type: object,null",
    "estimatedCostUsd": 0,
    "createdAt": "2024-07-29T15:51:28.071Z",
    "updatedAt": "2024-07-29T15:51:28.071Z"
    }