List Flow Tasks

Retrieves the scheduled flow run tasks on a site. A flow run task is the schedule for a Tableau Prep flow — when and how often it is configured to run — not a record of past executions. For run history, use List Flow Runs.

Each task includes the target flow (flow.id, flow.name), the schedule (frequency, next run time, state), and the task id used to trigger an on-demand run.

APIs called

• Get Flow Run Tasks

Required Tableau API scopes

When the MCP server authenticates with OAuth (connected-app JWT), this tool requests:

• tableau:flow_tasks:read
• tableau:mcp_site_settings:read

The tableau:flow_tasks:read scope was added in Tableau Cloud December 2025 / Server 2025.3. See OAuth configuration for how scopes are negotiated.

Caller-role visibility

• Non-admin callers — Tableau returns the scheduled tasks only for flows the caller owns.
• Server / site-admin callers — Tableau returns every scheduled flow task on the site.

This tool is not admin-gated; any authenticated user can call it and will see the tasks Tableau exposes to them.

Optional arguments

filter

A client-side filter string in the format field:operator:value. Multiple expressions are combined with a comma using a logical AND. The Tableau REST API does not support server-side filtering for this endpoint, so all tasks are fetched and filtered in-process.

Field	Type	Operators	Example
id	string	eq, in	id:eq:1bff10bb-57ae-43df-8774-a86d14aef432
type	string	eq, in	type:eq:RunFlowTask
priority	number	eq, gt, gte, lt, lte	priority:gte:5
consecutiveFailedCount	number	eq, gt, gte, lt, lte	consecutiveFailedCount:gt:0
flow.id	string	eq, in	flow.id:eq:8a320dca-9151-41ea-8474-...
flow.name	string	eq, in	flow.name:eq:Daily Sales
schedule.id	string	eq, in	schedule.id:eq:36d6fab2-2a0a-...
schedule.name	string	eq, in	schedule.name:eq:Daily Refresh
schedule.state	string	eq, in	schedule.state:eq:Active
schedule.frequency	string	eq, in	schedule.frequency:eq:Daily
schedule.nextRunAt	string (ISO 8601)	eq, gt, gte, lt, lte	schedule.nextRunAt:lt:2026-05-25T00:00:00Z
schedule.createdAt	string (ISO 8601)	eq, gt, gte, lt, lte	schedule.createdAt:gte:2026-01-01T00:00:00Z
schedule.updatedAt	string (ISO 8601)	eq, gt, gte, lt, lte	schedule.updatedAt:gte:2026-05-01T00:00:00Z

The in operator accepts a bracket/comma list (e.g. schedule.state:in:[Active,Suspended]), matching every other list-* tool; the legacy pipe-delimited form (schedule.state:in:Active|Suspended) is also accepted. Date-time values must be full ISO 8601 (e.g. 2026-05-25T00:00:00Z).

Examples:

• schedule.frequency:eq:Daily,schedule.state:eq:Active
• consecutiveFailedCount:gt:0 (schedules that are failing repeatedly)

---

pageSize

The maximum number of results per page, applied client-side after filtering.

limit

The maximum total number of tasks to return, applied client-side after filtering.

Response-size guidance

The Tableau "Get Flow Run Tasks" endpoint has no server-side filtering or pagination, so the tool fetches every scheduled task on the site and then applies filter/limit in-process. On large sites that is a big, slow response, so the tool description steers the LLM toward the narrowest call that answers the question:

Question shape	Recommended arguments
"When does flow X run next?"	filter: "flow.id:eq:<uuid>"
"Which schedules are failing?"	filter: "consecutiveFailedCount:gt:0"
"Which daily schedules are active?"	filter: "schedule.frequency:eq:Daily,schedule.state:eq:Active"
"Are any flows scheduled at all?"	a small limit (e.g. 10)

Bounded context (fail-closed)

A flow run task carries the flow's id and name but no project or tag, so when the server is restricted to an INCLUDE_PROJECT_IDS or INCLUDE_TAGS bounded context this tool cannot prove a task's flow belongs to the allowed set. Rather than risk leaking schedules for flows outside the allow-list, it returns no tasks and explains why.

Result info

The response is an object { flowTasks: [...], mcp: { resultInfo } }. The flowTasks array holds one record per scheduled task; mcp.resultInfo is always present and reports whether that array is complete:

• returnedCount — number of tasks in flowTasks.
• totalAvailable — number of tasks matching the filter before any limit. Always exact here, because the endpoint returns every task and filtering/limiting happen in-process.
• truncated — true when a limit/pageSize or an admin MAX_RESULT_LIMIT cut the list short.
• truncationReason — requested-limit (caller's limit/pageSize) or admin-cap (site limit), present only when truncated is true.

Example result

{
  "flowTasks": [
    {
      "id": "1bff10bb-57ae-43df-8774-a86d14aef432",
      "priority": 50,
      "consecutiveFailedCount": 2,
      "type": "RunFlowTask",
      "schedule": {
        "id": "36d6fab2-2a0a-432e-b464-9fe4229a9937",
        "name": "Every 2 Minutes",
        "state": "Active",
        "priority": 50,
        "createdAt": "2018-11-08T21:57:49Z",
        "updatedAt": "2018-11-09T17:30:08Z",
        "type": "Flow",
        "frequency": "Hourly",
        "nextRunAt": "2018-11-09T17:32:00Z"
      },
      "flow": {
        "id": "8a320dca-9151-41ea-8474-a0bb71961cc0",
        "name": "allUseCaseTFLX2"
      }
    }
  ],
  "mcp": { "resultInfo": { "returnedCount": 1, "truncated": false, "totalAvailable": 1 } }
}