NEnetxms
AI Operators

Modify AI operator instance

Changed on

Partially update an AI operator instance; only the fields present in the request body are changed.

patch/v1/ai/operators/{operator-id}

Request

  • Base URL
    http://127.0.0.1:8000
  • URL
    http://127.0.0.1:8000/v1/ai/operators/{operator-id}
  • Auth
    HTTP bearer

Path parameters

operator-idinteger required

Request body

namestring

Instance name (required on creation)

descriptionstring

Instance description

ownerUserIdinteger

Owner user account the instance executes as. Must be an existing, enabled user (not a group or the system account) with the AI assistant access right. Defaults to the built-in "ai-operator" account on creation.

enabledboolean

Enable or disable the instance

scopeFilterstring

Attention scope filter

modelSlotstring

AI model slot to use (empty = default)

minIntervalinteger

Minimum interval between executions in seconds (floor 60)

maxIntervalinteger

Maximum interval between executions in seconds (must be >= minInterval)

dailyTokenBudgetinteger

Daily LLM token budget (0 = unlimited)

personaPromptstring

Additional persona instructions

observationRetentionDaysinteger

Observation retention override in days (0 = server default)

observationMaxRecordsinteger

Observation record cap override (0 = server default)

instructionsstring

Standing instructions text (empty string clears)

instructionsLockedboolean

Lock standing instructions against updates by the operator

Response

Updated instance

idinteger

Unique instance identifier

namestring

Instance name

descriptionstring

Instance description

ownerUserIdinteger

ID of the owner user account. The instance executes with this account's rights: tool calls and standing check scripts see exactly what this user can access.

enabledboolean

True if the instance is enabled

executingboolean

True if an iteration is currently running

scopeFilterstring

Attention scope filter (object names or IDs)

modelSlotstring

AI model slot used for executions (empty = default)

minIntervalinteger

Minimum interval between executions in seconds

maxIntervalinteger

Maximum interval between executions in seconds

dailyTokenBudgetinteger

Daily LLM token budget (0 = unlimited)

tokensUsedTodayinteger

LLM tokens used within the current usage day (UTC)

personaPromptstring

Additional persona instructions appended to the system prompt

currentFocusstring

Current focus set by the operator itself

watchListstring

Watch list maintained by the operator itself

mementostring

State carried between iterations by the operator itself

observationRetentionDaysinteger

Observation retention override in days (0 = server default)

observationMaxRecordsinteger

Observation record cap override (0 = server default)

instructionsstring

Standing instructions authored by the operator itself, injected into its system prompt after the persona prompt

instructionsLockedboolean

True if an administrator locked the standing instructions; updates returned by the model are ignored while set

lastExecutionTimestring date-time

Time of the last completed execution

nextExecutionTimestring date-time

Scheduled time of the next execution

iterationinteger

Number of completed iterations

consecutiveFailuresinteger

Consecutive failed executions (resets on success and restart)

lastExplanationstring

Explanation returned by the last execution

createdstring date-time

Instance creation time

modifiedstring date-time

Last configuration change time

Changes