Task - Well Starter Kit

A Task represents a discrete unit of work within a workspace — it can be user-created, agent-created, or system-seeded as part of an onboarding or provider-scoring pipeline. Tasks belong to a single workspace, may be assigned to a Person, and may form a parent-child hierarchy (a task may have one parent task and many subtasks). They carry a rich status lifecycle, typed enums for executor and source, optional plan/step/scoring/subworkspace-candidate metadata in a JSONB field, and a polymorphic references array that links them to invoices, companies, people, documents, transactions, providers, or calendar months.

Naming	Value
Object	Task
Resource type (JSON:API `type`)	`task`
Collection / records root	`tasks`
REST base	`/v1/tasks`
Entity class	`Task`

API operations

Operation	Method & path	Status
List	`GET /v1/tasks`	✅ Implemented
Retrieve	`GET /v1/tasks/{id}`	✅ Implemented
Create	`POST /v1/tasks`	🟡 Planned
Update	`PATCH /v1/tasks/{id}`	🟡 Planned
Delete	`DELETE /v1/tasks/{id}`	🟡 Planned

Data model

Attributes

Field	Type	Required	Constraints	Allowed values	Description
task_id	string, UUID, system	✅ Yes	unique; generated by gen_random_uuid() on INSERT	—	Public stable identifier for the task. Never changes after creation.
title	string	✅ Yes	text; no length limit enforced at DB level	—	Short human-readable label for the task, shown in the UI and chat cards.
description	string	⚪ No	text; nullable	—	Optional longer prose explanation of what the task requires and why.
status	string (enum)	✅ Yes	default = ‘open’; backed by native PG enum task_status_enum; partial unique index uniq_tasks_active_provider_company fires only on status IN (‘open’,‘ice_log’,‘blocked’)	ice_log, open, in_progress, done, blocked, cancelled, archived	Lifecycle state of the task. ice_log = surfaced but not yet actionable; archived = closed but kept for audit.
executor_type	string (enum)	✅ Yes	NOT NULL; backed by native PG enum executor_type_enum	human, system	Whether the task is intended for a human actor or executed entirely by the system/agent pipeline.
source	string (enum)	✅ Yes	default = ‘user’; backed by native PG enum task_source_enum	user, agent	Whether the task was created by a human user or autonomously by an AI agent.
priority	string (enum)	⚪ No	nullable; default = ‘medium’; backed by native PG enum task_priority_enum	low, medium, high, critical	Urgency level of the task, used to sort and filter the task list in the UI.
confidence_score	number (decimal)	⚪ No	nullable; columnType numeric(3,2); range 0.00-1.00	—	Agent-assigned confidence that the task is relevant and actionable for this workspace. Higher means more certain.
token_reward	integer	⚪ No	nullable; integer	—	Gamification reward tokens awarded to the user upon completing this task. Sourced from the associated TaskTemplate or set by the pipeline.
due_date	datetime	⚪ No	nullable; no DB-level CHECK	—	Optional deadline after which the task is considered overdue. Displayed in the task card and used for sorting.
visible_date	datetime	⚪ No	nullable; no DB-level CHECK	—	Date before which the task should not be surfaced to the user (deferred visibility). Used by the scoring pipeline to schedule tasks for future monthly-close cycles.
references	jsonb (array of TaskReference objects)	✅ Yes	defaults to []; GIN index idx_tasks_references_gin (jsonb_path_ops) for fast dedup lookups	Array of { type: invoice\|company\|person\|document\|transaction\|provider\|month, id: string, label: string }	Polymorphic reference list linking the task to one or more domain entities. Drives context in the chat card and CTA routing. The first provider-typed entry is denormalized into provider_ref_id; the first company-typed entry into company_ref_id.
provider_ref_id	string	⚪ No	nullable; varchar(36); denormalized from references[0 where type=‘provider’].id; participates in partial unique index uniq_tasks_active_provider_company on (workspace_pk, provider_ref_id, company_ref_id) WHERE deleted_at IS NULL AND status IN (‘open’,‘ice_log’,‘blocked’) AND both ref_ids NOT NULL AND parent_task_pk IS NULL	—	Denormalized first provider reference id extracted from the references array. Populated at task creation by TaskService.createTask. Enables O(1) dedup guard preventing duplicate root scoring tasks for the same workspace x provider x company triplet.
company_ref_id	string	⚪ No	nullable; varchar(36); denormalized from references[0 where type=‘company’].id; co-participant in partial unique index uniq_tasks_active_provider_company	—	Denormalized first company reference id extracted from the references array. Populated at task creation by TaskService.createTask alongside provider_ref_id.
history	jsonb (array of TaskHistoryEntry objects)	✅ Yes	defaults to []	Array of { action: created\|status_changed\|assigned\|comment, at: ISO8601, by?: string, by_type: human\|system, from?: string, to?: string, detail?: string }	Append-only audit log of state transitions and comments on this task. Each entry records who did what and when.
plan_meta	jsonb (discriminated union on ‘kind’)	⚪ No	nullable. Partial unique index uq_tasks_subworkspace_candidate_dedup on (workspace_pk, plan_meta->>‘dedup_key’) WHERE deleted_at IS NULL AND plan_meta->>‘kind’ = ‘subworkspace-candidate’ AND plan_meta->>‘dedup_key’ IS NOT NULL.	kind: plan \| step \| scoring \| mail-connect \| connect-extension \| connect-tools-parent \| subworkspace-candidate	Optional metadata blob whose shape is discriminated by the ‘kind’ field. plan = agentic plan phases. step = individual step within a plan. scoring = provider-scoring signals and CTA route. mail-connect = connect-email fallback gate. connect-extension = Chrome extension install sub-task. connect-tools-parent = lazy parent for provider-connection subtasks. subworkspace-candidate = WAS-projection, detection signals, and dedup_key for child workspace discovery.
score	number (decimal)	⚪ No	nullable; columnType numeric(8,2)	—	Computed relevance or urgency score assigned by the scoring pipeline (0-1000+ scale). Used to order tasks in the UI surface.
done_at	datetime	⚪ No	nullable; no DB trigger; set by service layer	—	Timestamp when the task transitioned to status=done. Set by the service layer on completion.
conversation_thread_id	string, UUID	⚪ No	nullable; UuidType; soft reference only (no FK constraint)	—	UUID of the chat conversation thread associated with this task. Allows the AI to link back to the thread context. No FK cascade — deleting the conversation does not affect the task.
created_at	datetime, system	✅ Yes	set on INSERT via MikroORM onCreate lifecycle; not nullable	—	Timestamp when the task row was created.
updated_at	datetime, system	⚪ No	set on INSERT (onCreate) and updated on every UPDATE (onUpdate) via MikroORM lifecycle	—	Timestamp of the last modification to this task row.
deleted_at	datetime, system	⚪ No	nullable; soft-delete sentinel; all active queries filter deleted_at IS NULL; both partial unique indexes include deleted_at IS NULL in their WHERE clause so hard-deletes release dedup slots	—	Soft-delete timestamp. When set, the task is logically deleted and excluded from all active queries.

Relationships

Name	Type	Required	Description
workspace	to-one (workspace)	✅ Yes	The workspace this task belongs to. deleteRule = ‘cascade’ — deleting the workspace hard-deletes all its tasks. Every query must scope to this relationship. Composite index idx_tasks_workspace_status covers (workspace_pk, status) for efficient status-filtered list queries.
template	to-one (task_template)	⚪ No	Optional reference to the TaskTemplate that spawned or defines this task (title, markdown content, action type, token reward). deleteRule = ‘set null’ — deleting a template leaves tasks intact but unlinked. Template-less tasks are common for agent-generated or platform-seeded rows.
parent_task	to-one (task)	⚪ No	Self-referential parent for hierarchical task trees. A root task has parent_task = null. deleteRule = ‘set null’ — deleting a parent orphans its children rather than cascade-deleting them. The partial unique index uniq_tasks_active_provider_company only applies to root tasks (parent_task_pk IS NULL).
subtasks	to-many (task)	—	Inverse of parent_task. Collection of child tasks nested under this task. The composite composite_subtasks_list (display_type: relation_list, sort_proxy: subtasks_aggregate.min.created_at) in composites.yml surfaces subtask task_id, title, status, and due_date as a relation_list cell in the records table.
assigned_to	to-one (people)	⚪ No	The Person to whom the task is assigned for action. deleteRule = ‘set null’. The composite assigned_to.composite_avatar_fullname (display_type: people_avatar_name) in composites.yml surfaces person_id, avatar URL, and full_name as a composite cell in the records table.
created_by	to-one (people)	⚪ No	The Person who created the task, when created by a human. Null for agent-created or system-seeded tasks. deleteRule = ‘set null’.

System-computed

task_id is generated by gen_random_uuid() at INSERT via MikroORM defaultRaw; unique constraint enforced at DB level.
created_at is set by MikroORM onCreate lifecycle hook to new Date(); never subsequently updated.
updated_at is set by MikroORM onCreate and refreshed on every UPDATE via onUpdate lifecycle hook.
deleted_at is the soft-delete sentinel. All active queries must filter deleted_at IS NULL. Both partial unique indexes (uniq_tasks_active_provider_company and uq_tasks_subworkspace_candidate_dedup) include deleted_at IS NULL so hard-deleting a row releases its dedup slot.
status defaults to ‘open’ (TaskStatusEnum.OPEN) at entity construction; the native PG enum task_status_enum enforces allowed values at the DB level.
source defaults to ‘user’ (TaskSourceEnum.USER); overridden to ‘agent’ by the scoring and onboarding pipelines.
priority defaults to ‘medium’ (TaskPriorityEnum.MEDIUM); nullable so it can be intentionally unset.
references defaults to [] (empty array). The first provider-typed entry is extracted to provider_ref_id and the first company-typed entry to company_ref_id by TaskService.createTask. These two denormalized columns must stay in sync with the references array via service-layer discipline (no DB trigger enforces this).
history defaults to [] (empty array). Entries are appended by the service layer on every state transition (status_changed, assigned, comment). No DB trigger; the service is the sole writer.
Partial unique index uniq_tasks_active_provider_company on (workspace_pk, provider_ref_id, company_ref_id) WHERE deleted_at IS NULL AND status IN (‘open’,‘ice_log’,‘blocked’) AND provider_ref_id IS NOT NULL AND company_ref_id IS NOT NULL AND parent_task_pk IS NULL. Closes the concurrent-worker dedup race for root scoring tasks. Sub-tasks (parent_task_pk IS NOT NULL) are excluded by design.
Partial unique index uq_tasks_subworkspace_candidate_dedup on (workspace_pk, (plan_meta->>‘dedup_key’)) WHERE deleted_at IS NULL AND plan_meta->>‘kind’ = ‘subworkspace-candidate’ AND plan_meta->>‘dedup_key’ IS NOT NULL. Prevents re-emission of duplicate subworkspace-candidate tasks for the same canonical identity. CANCELLED rows are still covered; only hard-deleted rows release the slot.
Composite index idx_tasks_workspace_status on (workspace_pk, status) for efficient status-filtered workspace list queries.
GIN index idx_tasks_references_gin on references JSONB column (jsonb_path_ops) for fast reference-type filter and dedup lookups.
plan_meta is a discriminated-union JSONB blob. The ‘kind’ discriminant determines the valid shape. The rendering layer narrows on plan_meta.kind to select the correct chat card or CTA component.
done_at is set by the service layer when transitioning status to ‘done’; it is not a DB-generated column.
conversation_thread_id is a soft UUID reference to a ChatConversation; there is no FK constraint so deleting the conversation does not affect the task.

Example

{
  "data": {
    "type": "task",
    "id": "d4f1c7b2-3a9e-4f08-91c3-8e20b5f6a012",
    "attributes": {
      "title": "Connect Stripe to import missing invoices",
      "description": "We detected 14 open transactions from Stripe that have no matching invoice. Connect the Stripe provider to auto-import them.",
      "status": "open",
      "executor_type": "human",
      "source": "agent",
      "priority": "high",
      "confidence_score": "0.88",
      "token_reward": "50",
      "due_date": "2026-06-15T00:00:00.000Z",
      "visible_date": "2026-06-01T00:00:00.000Z",
      "score": "940.00",
      "done_at": null,
      "conversation_thread_id": null,
      "provider_ref_id": "a3f8c1d2-1111-4b0c-9abc-000000000001",
      "company_ref_id": "b7e2a5c4-2222-4c1d-8def-000000000002",
      "references": [
        { "type": "provider", "id": "a3f8c1d2-1111-4b0c-9abc-000000000001", "label": "Stripe" },
        { "type": "company",  "id": "b7e2a5c4-2222-4c1d-8def-000000000002", "label": "Stripe Payments Europe Ltd" }
      ],
      "history": [
        { "action": "created", "at": "2026-06-02T08:30:00.000Z", "by": null, "by_type": "system" },
        { "action": "assigned", "at": "2026-06-02T09:00:00.000Z", "by": "usr_human_01", "by_type": "human", "to": "Alice Martin" }
      ],
      "plan_meta": {
        "kind": "scoring",
        "signals": { "open_transactions": 14, "total_amount_eur": 12800, "dominant_currency": "EUR" },
        "route": { "action_type": "connect_provider" }
      },
      "created_at": "2026-06-02T08:30:00.000Z",
      "updated_at": "2026-06-02T09:00:00.000Z",
      "deleted_at": null
    },
    "relationships": {
      "workspace": { "data": { "type": "workspace", "id": "9f3a4b7c-0001-4e2a-b000-aabbccddeeff" } },
      "template": { "data": null },
      "parent_task": { "data": null },
      "subtasks": { "data": [] },
      "assigned_to": { "data": { "type": "people", "id": "c8d1f9e0-3333-4d2e-9e00-000000000003" } },
      "created_by": { "data": null }
    }
  }
}

_{Source: apps/api/src/database/entities/Task.ts · domain: workspace · tier: Activity}

Documentation Index

​API operations

​Data model

​Attributes

​Relationships

​System-computed

​Example