Skip to content

ark_job_submit

Run a long-running tool as a background job and return a job_id immediately.

Required for tools marked as requiring background execution (e.g. media_upload, vod_*); no MCP task support needed. Poll ark_job_get for the tool’s result. Do not auto-retry a timeout or disconnect: the job may already be queued.

Requirement Value
Registration Always registered; no provider credentials required.
Execution Foreground only
Required permission No fixed per-tool scope; HTTP authentication and target-tool authorization still apply.

This site lists the full configured capability surface, not the tools enabled on your particular server. Use model discovery and job discovery to check your connection. Credentials are server configuration, never tool arguments.

Ask your connected AI client to use ark_job_submit for the operation you want. The client constructs the arguments and handles tool calls. Review its proposed operation before approving anything billable. For a non-billable introduction, start with discovery without generation.

Call this tool directly. It is not a background-job target.

Match the input fields below and the full JSON Schema. See the worked examples for this capability and Background Execution for job/result handling.

The MCP arguments object wraps these fields as {"input": {...}}.

Field Type Required Description and constraints
arguments object Yes Full arguments object for the target tool, as in its input schema (usually {‘input’: {…}}).
tool_name string Yes Tool to run in the background; must be listed by ark_job_capabilities.; minLength: 1; maxLength: 128
Field Type Required Description and constraints
created_at string Yes ISO-8601 timestamp when the job was created.; format: “date-time”
job_id string Yes Server-generated Ark background job ID. This is distinct from any provider task ID returned inside the completed target result.; minLength: 20; maxLength: 128; pattern: ”^[A-Za-z0-9_-]+$”
poll_after_ms integer Yes Recommended delay in milliseconds before polling this job.; minimum: 0
status “working” No Initial job status after durable acceptance.; Default: “working”
target_tool string Yes Task-enabled Ark MCP tool executing in the background.
ttl_ms integer or null Yes Result retention duration in milliseconds, or null when unlimited.

Field types named above refer to these shared definitions. Optional/null fields, enum choices and bounds are retained in the complete schemas below. Schema validation is followed by runtime capability and security checks.

BackgroundJobSubmitInput
Field Type Required Description and constraints
arguments object Yes Full arguments object for the target tool, as in its input schema (usually {‘input’: {…}}).
tool_name string Yes Tool to run in the background; must be listed by ark_job_capabilities.; minLength: 1; maxLength: 128
JsonValue

No fields. Pass {}.

Input schema, including the MCP wrapper
{
"$defs": {
"BackgroundJobSubmitInput": {
"description": "Input model for ``ark_job_submit``.",
"properties": {
"arguments": {
"additionalProperties": {
"$ref": "#/$defs/JsonValue"
},
"description": "Full arguments object for the target tool, as in its input schema (usually {'input': {...}}).",
"type": "object"
},
"tool_name": {
"description": "Tool to run in the background; must be listed by ark_job_capabilities.",
"maxLength": 128,
"minLength": 1,
"type": "string"
}
},
"required": [
"tool_name",
"arguments"
],
"type": "object"
},
"JsonValue": {}
},
"additionalProperties": false,
"properties": {
"input": {
"$ref": "#/$defs/BackgroundJobSubmitInput"
}
},
"required": [
"input"
],
"type": "object"
}
Output schema, including nested definitions
{
"description": "Handle returned immediately after a background job is durably accepted.",
"properties": {
"created_at": {
"description": "ISO-8601 timestamp when the job was created.",
"format": "date-time",
"title": "Created At",
"type": "string"
},
"job_id": {
"description": "Server-generated Ark background job ID. This is distinct from any provider task ID returned inside the completed target result.",
"maxLength": 128,
"minLength": 20,
"pattern": "^[A-Za-z0-9_-]+$",
"title": "Job Id",
"type": "string"
},
"poll_after_ms": {
"description": "Recommended delay in milliseconds before polling this job.",
"minimum": 0,
"title": "Poll After Ms",
"type": "integer"
},
"status": {
"const": "working",
"default": "working",
"description": "Initial job status after durable acceptance.",
"title": "Status",
"type": "string"
},
"target_tool": {
"description": "Task-enabled Ark MCP tool executing in the background.",
"title": "Target Tool",
"type": "string"
},
"ttl_ms": {
"anyOf": [
{
"minimum": 0,
"type": "integer"
},
{
"type": "null"
}
],
"description": "Result retention duration in milliseconds, or null when unlimited.",
"title": "Ttl Ms"
}
},
"required": [
"job_id",
"target_tool",
"created_at",
"ttl_ms",
"poll_after_ms"
],
"title": "BackgroundJobAccepted",
"type": "object"
}
  • Missing tool: check provider credentials and feature flags, then restart the connection.
  • Invalid arguments: keep the correct wrapper and follow current model limits from discovery. A valid schema alone does not prove the chosen model supports the combination.
  • A job completes with an error: inspect the original tool result, not only the outer job status. Never automatically resubmit potentially billable work.
  • Local output writes from jobs require server-configured OUTPUT_ROOTS; HTTP cannot write to a client’s local disk.

See configuration, background execution, and artifact delivery.