Skip to main content
Glama

create_subtask

Creates a subtask under an existing parent task with title, assignee, priority, due date, and labels. Sets initial status from status_slug or the project default.

Instructions

Create a subtask under a parent task. Set status_slug for initial status (defaults to the project's default status, NOT the parent's status).

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "parent_task_id": "<parent_task_id>",
      "title": "<title>"
    },
    "name": "create_subtask"
  }
}

Input Schema

NameRequiredDescriptionDefault
titleYesSubtask title.
labelsNoLabels for the subtask.
due_dateNoDue date, RFC3339 (e.g. 2026-08-10T12:00:00Z).
priorityNoPriority: urgent, high, medium, low, none.medium
assignee_idNoAgent or user ID to assign the subtask to. Defaults to the creator if omitted.
descriptionNoSubtask description.
start_afterNoHold the subtask out of the queue until this moment, RFC3339 with Z or an offset (2026-11-01T09:00:00+03:00). From an agent it must be the published time of an external event, given with start_after_evidence; not a retry or poll interval. Independent of due_date.
status_slugNoStatus slug (e.g. 'todo'). Uses project default if omitted.
assignee_typeNoAssignee type: agent, user, or unassigned.
custom_fieldsNoCustom field values, keyed by field slug.
parent_task_idYesParent task ID.
estimated_hoursNoEstimated hours.
start_after_originNoOptional, only agent_evidence. The server derives the origin from the caller; a person's date needs no evidence.
start_after_evidenceNoRequired by the server for an agent's start_after: {event_ref, source, why}, the same rules as wait_for on=time.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed3 schema fields changedv0.1.7
    • changedInput schema / properties / start_after / description
      Previous value: -"Don't surface/feed this subtask before this RFC3339 timestamp. Independent of due_date."New value: +"Hold the subtask out of the queue until this moment, RFC3339 with Z or an offset (2026-11-01T09:00:00+03:00). From an agent it must be the published time of an external event, given with start_after_evidence; not a retry or poll interval. Independent of due_date."
    • addedInput schema / properties / start_after_evidence
      Added value: +{
      +  "description": "Required by the server for an agent's start_after: {event_ref, source, why}, the same rules as wait_for on=time.",
      +  "properties": {
      +    "event_ref": {
      +      "description": "The external event the moment belongs to, e.g. nightly-export (max 500 chars).",
      +      "type": "string"
      +    },
      +    "source": {
      +      "description": "Where the event's time is published: an https:// URL, comment:<uuid> or artifact:<uuid> written by someone else.",
      +      "type": "string"
      +    },
      +    "why": {
      +      "description": "What happens at start_after and why not earlier, 10 to 500 chars.",
      +      "type": "string"
      +    }
      +  },
      +  "type": "object"
      +}
    • addedInput schema / properties / start_after_origin
      Added value: +{
      +  "description": "Optional, only agent_evidence. The server derives the origin from the caller; a person's date needs no evidence.",
      +  "type": "string"
      +}
  2. First observedv0.1.1

TDQS

A3.9/5.0
Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already declare this is a non-read-only, non-destructive, non-idempotent, closed-world create. The description adds a genuinely non-obvious behavioral fact beyond them: an omitted status_slug resolves to the project default and explicitly not the parent's status, which is exactly the assumption an agent would otherwise get wrong. It still says nothing about permission requirements or what the call returns, but the annotations plus schema carry most of that.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Two sentences, front-loaded with the action and followed by the single most error-prone detail. Nothing is redundant with the schema and no sentence is wasted.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a create tool with 14 parameters, nested evidence objects and full schema coverage, the description covers the core action and the one surprising default, which is a reasonable division of labor. The remaining gap is routing guidance against create_task versus create_subtask, which the description leaves implicit.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so the baseline is 3, and the description earns a bump by clarifying the fallback semantics of status_slug beyond the schema's terse 'Uses project default if omitted' – it rules out inheritance from the parent. It does not touch the other 13 parameters (start_after evidence rules, assignee defaults), but those are fully documented in the schema.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb and resource ('Create a subtask under a parent task'), which immediately separates it from a plain task-creation tool and signals the required parent_task_id. It does not explicitly name create_task as the non-subtask alternative, so it stops short of full sibling differentiation.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines3/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Usage is implied by 'under a parent task' and by the required parent_task_id, so an agent can infer this is the subtask path rather than the top-level task path. There is no explicit when-to-use/when-not guidance, no mention of create_task or create_recurring_task, and no stated prerequisites or permissions.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.