protocol.mdx

---
title: Protocol schema — API Protocol reference
navTitle: Protocol
description: "Protocol schemas of the ObjectStack API Protocol: AiAgentCapabilities and 143 more — each property with its type, default and a TypeScript example."
---

{/* ⚠️  AUTO-GENERATED — DO NOT EDIT. Run build-docs.ts to regenerate. Hand-written docs live in the module folders under content/docs/. */}

<Callout type="info">
**Source:** `packages/spec/src/api/protocol.zod.ts`
</Callout>

## TypeScript Usage

```typescript
import { AiAgentCapabilitiesSchema, AiAgentChatRequestSchema, AiAgentSummarySchema, AiAgentsResponseSchema, AiChatRequestSchema, AiChatResponseSchema, AiCompleteRequestSchema, AiConversationSchema, AiMessageSchema, AiModelsResponseSchema, AiPendingActionSchema, AiPendingActionStatusSchema, AiStreamChunkSchema, ApproveAiPendingActionResponseSchema, AuditMetaItemRequestSchema, AuditMetaItemResponseSchema, AutomationActionsResponseSchema, AutomationTriggerRequestSchema, AutomationTriggerResponseSchema, BatchDataRequestSchema, BatchDataResponseSchema, CheckPermissionRequestSchema, CheckPermissionResponseSchema, CloneDataResponseSchema, CreateAiConversationRequestSchema, CreateDataRequestSchema, CreateDataResponseSchema, CreateManyDataRequestSchema, CreateManyDataResponseSchema, DeleteDataRequestSchema, DeleteDataResponseSchema, DeleteManyDataRequestSchema, DeleteManyDataResponseSchema, DeleteMetaItemRequestSchema, DeleteMetaItemResponseSchema, DiffMetaItemResponseSchema, DisablePackageRequestSchema, DisablePackageResponseSchema, EnablePackageRequestSchema, EnablePackageResponseSchema, FindDataRequestSchema, FindDataResponseSchema, FindReferencesToMetaResponseSchema, FlowRegistrationSchema, FlowRegistrationRefusalSchema, GetDataRequestSchema, GetDataResponseSchema, GetDiscoveryRequestSchema, GetDiscoveryResponseSchema, GetEffectivePermissionsRequestSchema, GetEffectivePermissionsResponseSchema, GetFieldLabelsRequestSchema, GetFieldLabelsResponseSchema, GetLocalesRequestSchema, GetLocalesResponseSchema, GetMetaDiagnosticsResponseSchema, GetMetaItemCachedRequestSchema, GetMetaItemCachedResponseSchema, GetMetaItemLayeredRequestSchema, GetMetaItemLayeredResponseSchema, GetMetaItemRequestSchema, GetMetaItemResponseSchema, GetMetaItemsRequestSchema, GetMetaItemsResponseSchema, GetMetaTypesRequestSchema, GetMetaTypesResponseSchema, GetNotificationPreferencesRequestSchema, GetNotificationPreferencesResponseSchema, GetObjectPermissionsRequestSchema, GetObjectPermissionsResponseSchema, GetPackageRequestSchema, GetPackageResponseSchema, GetPresenceRequestSchema, GetPresenceResponseSchema, GetPublishedMetaItemResponseSchema, GetRecordFieldAccessResponseSchema, GetTranslationsRequestSchema, GetTranslationsResponseSchema, GetUiViewRequestSchema, GetUiViewResponseSchema, HistoryMetaItemRequestSchema, HistoryMetaItemResponseSchema, HttpFindQueryParamsSchema, InstallPackageRequestSchema, InstallPackageResponseSchema, ListAiConversationsRequestSchema, ListAiConversationsResponseSchema, ListAiPendingActionsRequestSchema, ListAiPendingActionsResponseSchema, ListDraftsResponseSchema, ListNotificationsRequestSchema, ListNotificationsResponseSchema, ListPackagesRequestSchema, ListPackagesResponseSchema, MarkAllNotificationsReadRequestSchema, MarkAllNotificationsReadResponseSchema, MarkNotificationsReadRequestSchema, MarkNotificationsReadResponseSchema, MetadataAuthoringClientSchema, MetadataAuthoringSurfaceSchema, MetadataConflictErrorSchema, NotificationSchema, NotificationPreferencesSchema, PublicFormSubmitResponseSchema, PublishMetaItemRequestSchema, PublishMetaItemResponseSchema, PublishPackageDraftsRequestSchema, PublishPackageDraftsResponseSchema, RealtimeConnectRequestSchema, RealtimeConnectResponseSchema, RealtimeDisconnectRequestSchema, RealtimeDisconnectResponseSchema, RealtimeSubscribeRequestSchema, RealtimeSubscribeResponseSchema, RealtimeUnsubscribeRequestSchema, RealtimeUnsubscribeResponseSchema, RegisterDeviceRequestSchema, RegisterDeviceResponseSchema, RejectAiPendingActionResponseSchema, RollbackMetaItemResponseSchema, RuntimeAuthoringIssueSchema, SaveMetaItemRequestSchema, SaveMetaItemResponseSchema, SearchAllHitSchema, SearchAllPageHitSchema, SearchAllResponseSchema, SeedAppliedIdListSchema, SeedAppliedLandedSchema, SetPresenceRequestSchema, SetPresenceResponseSchema, UninstallPackageRequestSchema, UninstallPackageResponseSchema, UnregisterDeviceRequestSchema, UnregisterDeviceResponseSchema, UpdateAiConversationRequestSchema, UpdateDataRequestSchema, UpdateDataResponseSchema, UpdateManyDataRequestSchema, UpdateManyDataResponseSchema, UpdateNotificationPreferencesRequestSchema, UpdateNotificationPreferencesResponseSchema, ValidateDataIssueSchema, ValidateDataRequestSchema, ValidateDataResponseSchema } from '@objectstack/spec/api';
import type { AiAgentCapabilities, AiAgentChatRequest, AiAgentSummary, AiAgentsResponse, AiChatRequest, AiChatResponse, AiCompleteRequest, AiConversation, AiMessage, AiModelsResponse, AiPendingAction, AiPendingActionStatus, AiStreamChunk, ApproveAiPendingActionResponse, AuditMetaItemRequest, AuditMetaItemResponse, AutomationActionsResponse, AutomationTriggerRequest, AutomationTriggerResponse, BatchDataRequest, BatchDataResponse, CheckPermissionRequest, CheckPermissionResponse, CloneDataResponse, CreateAiConversationRequest, CreateDataRequest, CreateDataResponse, CreateManyDataRequest, CreateManyDataResponse, DeleteDataRequest, DeleteDataResponse, DeleteManyDataRequest, DeleteManyDataResponse, DeleteMetaItemRequest, DeleteMetaItemResponse, DiffMetaItemResponse, DisablePackageRequest, DisablePackageResponse, EnablePackageRequest, EnablePackageResponse, FindDataRequest, FindDataResponse, FindReferencesToMetaResponse, FlowRegistration, FlowRegistrationRefusal, GetDataRequest, GetDataResponse, GetDiscoveryRequest, GetDiscoveryResponse, GetEffectivePermissionsRequest, GetEffectivePermissionsResponse, GetFieldLabelsRequest, GetFieldLabelsResponse, GetLocalesRequest, GetLocalesResponse, GetMetaDiagnosticsResponse, GetMetaItemCachedRequest, GetMetaItemCachedResponse, GetMetaItemLayeredRequest, GetMetaItemLayeredResponse, GetMetaItemRequest, GetMetaItemResponse, GetMetaItemsRequest, GetMetaItemsResponse, GetMetaTypesRequest, GetMetaTypesResponse, GetNotificationPreferencesRequest, GetNotificationPreferencesResponse, GetObjectPermissionsRequest, GetObjectPermissionsResponse, GetPackageRequest, GetPackageResponse, GetPresenceRequest, GetPresenceResponse, GetPublishedMetaItemResponse, GetRecordFieldAccessResponse, GetTranslationsRequest, GetTranslationsResponse, GetUiViewRequest, GetUiViewResponse, HistoryMetaItemRequest, HistoryMetaItemResponse, InstallPackageRequest, InstallPackageResponse, ListAiConversationsRequest, ListAiConversationsResponse, ListAiPendingActionsRequest, ListAiPendingActionsResponse, ListDraftsResponse, ListNotificationsRequest, ListNotificationsResponse, ListPackagesRequest, ListPackagesResponse, MarkAllNotificationsReadRequest, MarkAllNotificationsReadResponse, MarkNotificationsReadRequest, MarkNotificationsReadResponse, MetadataAuthoringClient, MetadataAuthoringSurface, MetadataConflictError, Notification, NotificationPreferences, PublicFormSubmitResponse, PublishMetaItemRequest, PublishMetaItemResponse, PublishPackageDraftsRequest, PublishPackageDraftsResponse, RealtimeConnectRequest, RealtimeConnectResponse, RealtimeDisconnectRequest, RealtimeDisconnectResponse, RealtimeSubscribeRequest, RealtimeSubscribeResponse, RealtimeUnsubscribeRequest, RealtimeUnsubscribeResponse, RegisterDeviceRequest, RegisterDeviceResponse, RejectAiPendingActionResponse, RollbackMetaItemResponse, RuntimeAuthoringIssue, SaveMetaItemRequest, SaveMetaItemResponse, SearchAllHit, SearchAllPageHit, SearchAllResponse, SeedAppliedIdList, SeedAppliedLanded, SetPresenceRequest, SetPresenceResponse, UninstallPackageRequest, UninstallPackageResponse, UnregisterDeviceRequest, UnregisterDeviceResponse, UpdateAiConversationRequest, UpdateDataRequest, UpdateDataResponse, UpdateManyDataRequest, UpdateManyDataResponse, UpdateNotificationPreferencesRequest, UpdateNotificationPreferencesResponse, ValidateDataIssue, ValidateDataRequest, ValidateDataResponse } from '@objectstack/spec/api';

// Validate data
const result = AiAgentCapabilitiesSchema.parse(data);
```

---

## AiAgentCapabilities

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **authoring** | `boolean` | ✅ | Authors app metadata (objects/views/flows) |
| **canvas** | `boolean` | ✅ | Drives the Live Canvas split view (ADR-0037) |
| **debug** | `boolean` | ✅ | Exposes the build-doctor debug drawer |
| **resume** | `boolean` | ✅ | Turns resume durable multi-step runs (ADR-0013) |


---

## AiAgentChatRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **messages** | `({ role: Enum<'system' \| 'user' \| 'assistant' \| 'tool'>; content?: any; parts?: any[] } & Record<string, any>)[]` | ✅ | Conversation messages (at least one) |
| **context** | `Record<string, any>` | optional | Agent context (app, object, record, …) |
| **options** | `Record<string, any>` | optional | Request options (model, temperature, …) |

### Nested Shape: `AiAgentChatRequest.messages[number]`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **role** | `Enum<'system' \| 'user' \| 'assistant' \| 'tool'>` | ✅ | Message role |
| **content** | `any` | optional | Message content: a string, or an array of content parts |
| **parts** | `any[]` | optional | Vercel AI SDK v6 message parts (alternative to `content`) |


---

## AiAgentSummary

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **name** | `string` | ✅ | Agent name — the `:agentName` path segment |
| **label** | `string` | ✅ | Display label |
| **role** | `string` | ✅ | Agent role |
| **capabilities** | `{ authoring: boolean; canvas: boolean; debug: boolean; resume: boolean }` | ✅ | Capability set implied by the agent surface |

### Nested Shape: `AiAgentSummary.capabilities`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **authoring** | `boolean` | ✅ | Authors app metadata (objects/views/flows) |
| **canvas** | `boolean` | ✅ | Drives the Live Canvas split view (ADR-0037) |
| **debug** | `boolean` | ✅ | Exposes the build-doctor debug drawer |
| **resume** | `boolean` | ✅ | Turns resume durable multi-step runs (ADR-0013) |


---

## AiAgentsResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **agents** | `{ name: string; label: string; role: string; capabilities: object }[]` | ✅ | Agents this caller may chat with |

### Nested Shape: `AiAgentsResponse.agents[number]`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **name** | `string` | ✅ | Agent name — the `:agentName` path segment |
| **label** | `string` | ✅ | Display label |
| **role** | `string` | ✅ | Agent role |
| **capabilities** | `{ authoring: boolean; canvas: boolean; debug: boolean; resume: boolean }` | ✅ | Capability set implied by the agent surface |


---

## AiChatRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **messages** | `({ role: Enum<'system' \| 'user' \| 'assistant' \| 'tool'>; content?: any; parts?: any[] } & Record<string, any>)[]` | ✅ | Conversation messages (at least one) |
| **system** | `string` | optional | System prompt, prepended as a system message |
| **model** | `string` | optional | Model id override |
| **temperature** | `number` | optional | Sampling temperature |
| **maxTokens** | `integer` | optional | Maximum tokens to generate |
| **stream** | `boolean` | optional | false → JSON response; otherwise the UI Message Stream Protocol |
| **conversationId** | `string` | optional | Conversation to persist this turn into (auto-created when omitted) |
| **turnId** | `string` | optional | Stable per-turn idempotency key (ADR-0013 D1) |
| **options** | `Record<string, any>` | optional | Legacy nested request options |

### Nested Shape: `AiChatRequest.messages[number]`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **role** | `Enum<'system' \| 'user' \| 'assistant' \| 'tool'>` | ✅ | Message role |
| **content** | `any` | optional | Message content: a string, or an array of content parts |
| **parts** | `any[]` | optional | Vercel AI SDK v6 message parts (alternative to `content`) |


---

## AiChatResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **content** | `string` | ✅ | Generated text |
| **model** | `string` | optional | Model that produced it |
| **toolCalls** | `any[]` | optional | Tool calls the model requested (Vercel `ToolCallPart`) |
| **usage** | `{ promptTokens: number; completionTokens: number; totalTokens: number }` | optional | Token usage |
| **conversationId** | `string` | optional | Conversation the turn was persisted into |

### Nested Shape: `AiChatResponse.usage`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **promptTokens** | `number` | ✅ | Tokens consumed by the prompt |
| **completionTokens** | `number` | ✅ | Tokens generated |
| **totalTokens** | `number` | ✅ | prompt + completion |


---

## AiCompleteRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **prompt** | `string` | ✅ | Prompt text |
| **options** | `Record<string, any>` | optional | Request options (model, temperature, maxTokens, …) |


---

## AiConversation

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **id** | `string` | ✅ | Conversation id |
| **title** | `string` | optional | Title / summary |
| **agentId** | `string` | optional | Agent this conversation is bound to |
| **userId** | `string` | optional | Owning user |
| **messages** | `({ role: Enum<'system' \| 'user' \| 'assistant' \| 'tool'>; content?: any; parts?: any[] } & Record<string, any>)[]` | ✅ | Message history |
| **createdAt** | `string` | ✅ | Creation timestamp (ISO 8601) |
| **updatedAt** | `string` | ✅ | Last update timestamp (ISO 8601) |
| **metadata** | `Record<string, any>` | optional | Conversation metadata |

### Nested Shape: `AiConversation.messages[number]`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **role** | `Enum<'system' \| 'user' \| 'assistant' \| 'tool'>` | ✅ | Message role |
| **content** | `any` | optional | Message content: a string, or an array of content parts |
| **parts** | `any[]` | optional | Vercel AI SDK v6 message parts (alternative to `content`) |


---

## AiMessage

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **role** | `Enum<'system' \| 'user' \| 'assistant' \| 'tool'>` | ✅ | Message role |
| **content** | `any` | optional | Message content: a string, or an array of content parts |
| **parts** | `any[]` | optional | Vercel AI SDK v6 message parts (alternative to `content`) |


---

## AiModelsResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **models** | `(string \| { id: string; label: string; default: boolean })[]` | ✅ | Models this system offers |
| **defaultModel** | `string` | optional | Default model id, when the service reports one |

### Nested Shape: `AiModelsResponse.models[number]`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **id** | `string` | ✅ | Model id |
| **label** | `string` | ✅ | Display label for the picker |
| **default** | `boolean` | ✅ | Whether this is the environment default |


---

## AiPendingAction

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **id** | `string` | ✅ | Pending action id |
| **object_name** | `string` | ✅ | Object the action targets |
| **action_name** | `string` | ✅ | Action name |
| **tool_name** | `string` | ✅ | Tool that would execute it |
| **tool_input** | `string` | ✅ | Serialized tool input |
| **status** | `Enum<'pending' \| 'approved' \| 'executed' \| 'failed' \| 'rejected'>` | ✅ | Lifecycle status |
| **result** | `string` | optional | Serialized result, once executed |
| **error** | `string` | optional | Failure message, when status is failed |
| **rejection_reason** | `string` | optional | Reason given at rejection |
| **conversation_id** | `string` | optional | Conversation that proposed it |
| **message_id** | `string` | optional | Message that proposed it |
| **proposed_by** | `string` | optional | Actor that proposed it |
| **decided_by** | `string` | optional | Actor that approved or rejected it |
| **proposed_at** | `string` | ✅ | Proposal timestamp (ISO 8601) |
| **decided_at** | `string` | optional | Decision timestamp (ISO 8601) |


---

## AiPendingActionStatus

### Allowed Values

* `pending`
* `approved`
* `executed`
* `failed`
* `rejected`


---

## AiStreamChunk

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **type** | `string` | ✅ | Frame type (text-delta, tool-input-available, finish, error, …) |


---

## ApproveAiPendingActionResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **status** | `Enum<'executed' \| 'failed'>` | ✅ | Outcome of executing the approved action |
| **result** | `any` | optional | Tool result, when executed |
| **error** | `string` | optional | Failure reason, when failed |


---

## AuditMetaItemRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **type** | `string` | ✅ | Metadata type name |
| **name** | `string` | ✅ | Item name |
| **limit** | `number` | optional | Maximum events to return, newest first. The implementation clamps to [1, 500] and defaults to 100 — out-of-range values are clamped, never refused. |


---

## AuditMetaItemResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **events** | `{ id: any; occurredAt: string; actor: string; source: string \| null; … }[]` | ✅ | Recent protection-audit events for the item, newest first. See the schema-level note for what an empty array means — and what it never means. |

### Nested Shape: `AuditMetaItemResponse.events[number]`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **id** | `any` | ✅ | Row id of the audit event. Opaque to callers. |
| **occurredAt** | `string` | ✅ | When the attempt happened (ISO-8601 string). |
| **actor** | `string` | ✅ | Who attempted the operation. `system` when the row recorded no actor. |
| **source** | `string \| null` | ✅ | Which code path recorded the event (e.g. `protocol.deleteMetaItem`). `null` when the row recorded none. |
| **operation** | `Enum<'save' \| 'publish' \| 'rollback' \| 'delete' \| 'reset'>` | ✅ | Which metadata-protection door was attempted. |
| **outcome** | `Enum<'allowed' \| 'denied' \| 'forced'>` | ✅ | Whether the attempt went through, was refused, or overrode a lock (ADR-0010 §3.6). |
| **code** | `string` | ✅ | Machine-readable verdict code for the outcome (e.g. `item_locked`). Empty string when the row recorded none. |
| **lockState** | `Enum<'none' \| 'no-overlay' \| 'no-delete' \| 'full'> \| null` | ✅ | The lock verdict in force at the time of the attempt (ADR-0010 §3.3). `null` when no lock applied. |
| **lockOverridden** | `boolean` | ✅ | True when the attempt went through by overriding a lock (`outcome: "forced"` rows). |
| **requestId** | `string \| null` | ✅ | Correlation id of the originating request. `null` when the row recorded none. |
| **note** | `string \| null` | ✅ | Free-text note recorded with the event. `null` when the row recorded none. |


---

## AutomationActionsResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **actions** | `{ type: string; version: string; name: string; description?: string; … }[]` | ✅ | Registered action descriptors (built-in + plugin) |
| **total** | `integer` | ✅ | Number of descriptors returned (after any filters) |

### Nested Shape: `AutomationActionsResponse.actions[number]`

Canonical cross-paradigm action/node descriptor (ADR-0018)

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **type** | `string` | ✅ | Registry action/node type (matches the executor type) |
| **version** | `string` | ✅ | Executor version (semver) |
| **name** | `string` | ✅ | Display label (or i18n key) |
| **description** | `string` | optional | Action description |
| **icon** | `string` | optional | Icon id resolved by the designer |
| **category** | `Enum<'logic' \| 'data' \| 'io' \| 'human' \| 'control' \| 'custom'>` | optional (default: `"custom"`) | Palette category |
| **paradigms** | `Enum<'flow' \| 'approval'>[]` | optional (default: `["flow"]`) | Authoring surfaces that may offer this action |
| **configSchema** | `any` | optional | JSON Schema for the node config (drives the designer form; undeclared keys are rejected at registration) |
| **supportsPause** | `boolean` | optional (default: `false`) | Supports async pause/resume |
| **supportsCancellation** | `boolean` | optional (default: `false`) | Supports cancellation |
| **supportsRetry** | `boolean` | optional (default: `true`) | Supports retry on failure |
| **needsOutbox** | `boolean` | optional (default: `false`) | Dispatch via service-messaging outbox (retry/idempotency/dead-letter) |
| **isAsync** | `never` | optional | [REMOVED] `ActionDescriptor.isAsync` was removed in @objectstack/spec 17 (ADR-0049) — no execution path ever read it, so declaring it never made a node suspend and omitting it never stopped one. Delete the key. The live mechanism is two-part: an executor suspends by RETURNING `suspend: true` from `execute()`, and its descriptor must declare `supportsPause: true` (plus the `resumeAuthority` its pauses need) or the engine refuses that suspension. Declaring `isAsync: true` alongside `supportsPause: true` was always redundant; declaring it alone was always inert. |
| **handlerContract** | `Enum<'none' \| 'pure'>` | optional (default: `"none"`) | Effect contract for author-supplied code this action invokes: 'none' (invokes none) or 'pure' (must not write — it returns a value and the flow graph persists it) |
| **resumeAuthority** | `Enum<'any' \| 'service'>` | optional | Who may resume a run this node suspended: 'any' (the generic resume route) or 'service' (only the owning service, e.g. approvals). Carries no schema default so an omission stays observable — and an omission is fail-CLOSED at run time, equivalent to 'service': a pausing node whose pause is open to the generic route must declare 'any' explicitly |
| **maturity** | `Enum<'ga' \| 'beta' \| 'reserved'>` | optional (default: `"ga"`) | Runtime maturity: ga (shipped), beta, or reserved (contract only — designers grey this out) |
| **source** | `Enum<'builtin' \| 'plugin'>` | optional (default: `"plugin"`) | builtin = platform baseline; plugin = third-party contributed |
| **deprecated** | `boolean` | optional (default: `false`) | Deprecated alias kept for back-compat |
| **aliasOf** | `string` | optional | Canonical type this alias forwards to |


---

## AutomationTriggerRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **trigger** | `string` | ✅ |  |
| **payload** | `Record<string, any>` | ✅ |  |


---

## AutomationTriggerResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **success** | `boolean` | ✅ |  |
| **jobId** | `string` | optional |  |
| **result** | `any` | optional |  |


---

## BatchDataRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **object** | `string` | ✅ | Object name |
| **request** | `{ operation: Enum<'create' \| 'update' \| 'upsert' \| 'delete'>; records: object[]; options?: object }` | ✅ | Batch operation request |

### Nested Shape: `BatchDataRequest.request`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **operation** | `Enum<'create' \| 'update' \| 'upsert' \| 'delete'>` | ✅ | Type of batch operation |
| **records** | `{ id?: string; data?: Record<string, any>; externalId?: string }[]` | ✅ | Array of records to process (server caps the count — see batch.maxBatchSize) |
| **options** | `{ atomic?: boolean; returnRecords?: boolean; continueOnError?: boolean }` | optional | Batch operation options |


---

## BatchDataResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **success** | `boolean` | ✅ | Operation success status |
| **error** | `{ code: Enum<'VALIDATION_ERROR' \| 'INVALID_FIELD' \| 'MISSING_REQUIRED_FIELD' \| …>; declaredCode?: string; message: string; userMessage?: string; … }` | optional | Error details if success is false |
| **meta** | `{ timestamp: string; duration?: integer; requestId?: string; traceId?: string }` | optional | Response metadata |
| **operation** | `Enum<'create' \| 'update' \| 'upsert' \| 'delete'>` | optional | Operation type that was performed |
| **total** | `number` | ✅ | Total number of records in the batch |
| **succeeded** | `number` | ✅ | Number of records that succeeded |
| **failed** | `number` | ✅ | Number of records that failed |
| **results** | `{ id?: string; success: boolean; errors?: object[]; data?: Record<string, any>; … }[]` | ✅ | Detailed results for each record |

### Nested Shape: `BatchDataResponse.error`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **code** | `Enum<'VALIDATION_ERROR' \| 'INVALID_FIELD' \| 'MISSING_REQUIRED_FIELD' \| …>` | ✅ | Error code (e.g. VALIDATION_ERROR; StandardErrorCode ∪ the ledger the serving side registers — ERROR_CODE_LEDGER for framework packages) |
| **declaredCode** | `string` | optional | The producer-declared code, verbatim, when it is not a member of the closed `code` vocabulary — the open, author-authored channel (app-specific spellings; ADR-0112) |
| **message** | `string` | ✅ | Readable error message |
| **userMessage** | `string` | optional | Producer-marked user-facing refusal text, verbatim. Present exactly when the producer opted in at throw time; consumers render it to end users and keep their generic substitution for anything unmarked. Status-agnostic; never replaces `message`. |
| **refusal** | `true` | optional | Producer-declared: the 5xx this envelope carries is a deliberate refusal whose `message` is authored for the caller, so a boundary that reads the declaration keeps it verbatim (until the withhold arms read it, a declared refusal is still withheld). Absent (the default) on a declared fault, whose `message` is withheld from the body and logged for the operator; redundant on a 4xx. Presence is the declaration — `true` is the only value. |
| **category** | `string` | optional | Error category (e.g. validation, authorization) |
| **httpStatus** | `integer` | optional | HTTP status of the response carrying this error |
| **details** | `any` | optional | Additional error context (e.g. field validation errors) |
| **requestId** | `string` | optional | Request ID for tracking |

### Nested Shape: `BatchDataResponse.meta`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **timestamp** | `string` | ✅ |  |
| **duration** | `integer` | optional | Server-side processing duration in milliseconds |
| **requestId** | `string` | optional |  |
| **traceId** | `string` | optional |  |

### Nested Shape: `BatchDataResponse.results[number]`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **id** | `string` | optional | Record ID if operation succeeded |
| **success** | `boolean` | ✅ | Whether this record was processed successfully |
| **errors** | `{ code: Enum<'VALIDATION_ERROR' \| 'INVALID_FIELD' \| 'MISSING_REQUIRED_FIELD' \| …>; declaredCode?: string; message: string; userMessage?: string; … }[]` | optional | Array of errors if operation failed. Branch on `errors[0].code` — an atomic batch that rolled back marks rows that were written then undone with code ROLLED_BACK and rows never reached with NOT_ATTEMPTED, while the causal row keeps its own error. A NON-atomic batch that stopped (the `continueOnError: false` default) marks its un-attempted tail with the same NOT_ATTEMPTED code — rows before the failure stay written and keep reporting success, since nothing was rolled back. |
| **data** | `Record<string, any>` | optional | Full record data (if returnRecords=true) |
| **index** | `number` | optional | Index of the record in the request array |
| **droppedFields** | `{ object: string; fields: string[]; reason: Enum<'readonly' \| 'readonly_when' \| 'primary_key' \| 'computed' \| 'stored_expression'> }[]` | optional | Write-observability: caller-supplied fields LEGALLY stripped from THIS row before it was written — static `readonly` / TRUE `readonlyWhen` on update, or the in-engine static `readonly` strip on create. Per-row because a batch can drop different fields on different rows (`readonlyWhen` is record-state-dependent). Present ONLY when ≥1 field was dropped for this row; the row still succeeded (success unchanged). A single response header cannot say which row dropped what, so this body field stays the canonical per-row channel; REST also sends the `X-ObjectStack-Dropped-Fields` response header on the answer, naming the union of the rows' drops (each field and reason once). Optional — omit-when-empty keeps the shape backward-compatible. |


---

## CheckPermissionRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **object** | `string` | ✅ | Object name to check permissions for |
| **action** | `Enum<'create' \| 'read' \| 'edit' \| 'delete' \| 'transfer' \| 'restore' \| 'purge'>` | ✅ | Action to check |
| **recordId** | `string` | optional | Specific record ID (for record-level checks) |
| **field** | `string` | optional | Specific field name (for field-level checks) |


---

## CheckPermissionResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **allowed** | `boolean` | ✅ | Whether the action is permitted |
| **reason** | `string` | optional | Reason if denied |


---

## CloneDataResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **object** | `string` | ✅ | The object name. |
| **id** | `string` | ✅ | The ID of the newly created clone. |
| **sourceId** | `string` | ✅ | The ID of the record the clone was copied from. |
| **record** | `Record<string, any>` | ✅ | The created clone, including server-generated fields. Engine-owned values (injected system/audit columns, autonumbers, computed formula/summary fields) are re-derived by the insert path rather than copied from the source; caller-supplied `overrides` win over copied values. |
| **droppedFields** | `{ object: string; fields: string[]; reason: Enum<'readonly' \| 'readonly_when' \| 'primary_key' \| 'computed' \| 'stored_expression'> }[]` | optional | Write-observability: fields that were LEGALLY stripped before the clone was written — a non-system clone cannot seed a static `readonly` column, whether the value was COPIED from the source row or supplied through `overrides` (the strip runs inside `engine.insert`, after the `beforeInsert` hooks, `isSystem`-gated, exactly as on `createData`), so those keys are dropped and the field re-derives its default. Present ONLY when ≥1 field was dropped; the clone still succeeded without them (status/success semantics unchanged). REST additionally surfaces this as the `X-ObjectStack-Dropped-Fields` response header. Optional — omit-when-empty keeps the shape backward-compatible for existing clients. |
| **warnings** | `{ field: string; code: string; message: string; rule?: string; … }[]` | optional | The advisory validation-rule hits of THIS write: each `severity: 'warning'` or `'info'` rule whose verdict on the written record was "violated", with its `rule` name, `severity`, `field`, `code` and author-written `message` — the same element the validate-only preview reports in a row's `warnings`. Advisory rules never block, so the write succeeded; this is the advisory to show the person who made it. Present ONLY when at least one advisory rule hit; a write with no hit omits the key. Hits of a nested write a hook or a flow made inside this one are not included, and a rule that could not be evaluated is not a hit. |

### Nested Shape: `CloneDataResponse.droppedFields[number]`

A write-path strip event: caller-supplied fields legally dropped from the payload, or written with the engine's computed value in place of the caller's

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **object** | `string` | ✅ | Object the write targeted (resolved object name) |
| **fields** | `string[]` | ✅ | Caller-supplied field names whose value the engine did not store: removed from the write payload, or (`stored_expression`) written with the computed value in place of the caller's |
| **reason** | `Enum<'readonly' \| 'readonly_when' \| 'primary_key' \| 'computed' \| 'stored_expression'>` | ✅ | Why the fields were dropped: static readonly, a TRUE readonlyWhen predicate, the primary-key strip of a payload id the engine ruled is not an identifier, a computed (formula) field no driver has a column for, or a stored field whose declared expression the engine computes on every write, so the caller's different value was replaced by the computed one |

### Nested Shape: `CloneDataResponse.warnings[number]`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **field** | `string` | ✅ | The field the finding is about (`_record` for an object-level rule). |
| **code** | `string` | ✅ | Machine-readable finding code, e.g. `required`, `invalid_type`, `rule_violation`. |
| **message** | `string` | ✅ | Human-readable message — a validation rule's author-written text where one exists. |
| **rule** | `string` | optional | The `name` of the ADVISORY validation rule this finding reports. Present only on an advisory rule hit. |
| **severity** | `Enum<'warning' \| 'info'>` | optional | The advisory rule's declared severity, `warning` or `info`. Present only on an advisory rule hit. |


---

## CreateAiConversationRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **title** | `string` | optional | Initial title |
| **agentId** | `string` | optional | Agent to bind the conversation to |
| **metadata** | `Record<string, any>` | optional | Conversation metadata |


---

## CreateDataRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **object** | `string` | ✅ | The object name. |
| **data** | `Record<string, any>` | ✅ | The dictionary of field values to insert. |


---

## CreateDataResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **object** | `string` | ✅ | The object name. |
| **id** | `string` | ✅ | The ID of the newly created record. |
| **record** | `Record<string, any>` | ✅ | The created record, including server-generated fields (created_at, owner). |
| **droppedFields** | `{ object: string; fields: string[]; reason: Enum<'readonly' \| 'readonly_when' \| 'primary_key' \| 'computed' \| 'stored_expression'> }[]` | optional | Write-observability: caller-supplied fields that were LEGALLY stripped before the record was written — a non-system create cannot seed a static `readonly` column (the strip runs inside `engine.insert`, after the `beforeInsert` hooks, `isSystem`-gated), so those keys are dropped and the field re-derives its default. Present ONLY when ≥1 field was dropped; the create still succeeded without them (status/success semantics unchanged). REST additionally surfaces this as the `X-ObjectStack-Dropped-Fields` response header. Optional — omit-when-empty keeps the shape backward-compatible for existing clients. |
| **warnings** | `{ field: string; code: string; message: string; rule?: string; … }[]` | optional | The advisory validation-rule hits of THIS write: each `severity: 'warning'` or `'info'` rule whose verdict on the written record was "violated", with its `rule` name, `severity`, `field`, `code` and author-written `message` — the same element the validate-only preview reports in a row's `warnings`. Advisory rules never block, so the write succeeded; this is the advisory to show the person who made it. Present ONLY when at least one advisory rule hit; a write with no hit omits the key. Hits of a nested write a hook or a flow made inside this one are not included, and a rule that could not be evaluated is not a hit. |
| **withheldMentions** | `string[]` | optional | The ids of the users this write @mentioned whose mention notification was WITHHELD by the mention recipient gate, so they will not be notified: the gate's own verdict on THIS write, ids only, never the reason. Only a `sys_comment` write fills it, and only on the requester's own answer to the request that made the write; on an update, only a mention the update added is judged. The comment was saved with every mention; the requester is told here that the named users will not be notified. Response-only: never stored on the comment (a stored member would tell every reader who cannot read the thread). Present ONLY when at least one mention was withheld; otherwise the key is omitted. |
| **notice** | `string` | optional | One author-written message telling the person who made THIS write what the save set in motion (for example "Sent to your manager for approval"), set by a hook running on the write through the write's own `onNotice` listener (`WriteObservabilityOptions`; a hook body calls it as `ctx.notice(message)`), or by a `save_notice` node of a flow the write triggered. Info severity: it reports an outcome, never a problem, so it is distinct from `warnings` (the advisory validation-rule hits) and a client shows it with, or in place of, its own success message. One message: when more than one call sets it, the LAST one wins. Present ONLY when a hook set a non-empty message on a write that succeeded; otherwise the key is omitted. A nested write a hook or a flow made inside this one carries its own options, so its notice never reaches this answer. Response-only: never stored on the record. |

### Nested Shape: `CreateDataResponse.droppedFields[number]`

A write-path strip event: caller-supplied fields legally dropped from the payload, or written with the engine's computed value in place of the caller's

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **object** | `string` | ✅ | Object the write targeted (resolved object name) |
| **fields** | `string[]` | ✅ | Caller-supplied field names whose value the engine did not store: removed from the write payload, or (`stored_expression`) written with the computed value in place of the caller's |
| **reason** | `Enum<'readonly' \| 'readonly_when' \| 'primary_key' \| 'computed' \| 'stored_expression'>` | ✅ | Why the fields were dropped: static readonly, a TRUE readonlyWhen predicate, the primary-key strip of a payload id the engine ruled is not an identifier, a computed (formula) field no driver has a column for, or a stored field whose declared expression the engine computes on every write, so the caller's different value was replaced by the computed one |

### Nested Shape: `CreateDataResponse.warnings[number]`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **field** | `string` | ✅ | The field the finding is about (`_record` for an object-level rule). |
| **code** | `string` | ✅ | Machine-readable finding code, e.g. `required`, `invalid_type`, `rule_violation`. |
| **message** | `string` | ✅ | Human-readable message — a validation rule's author-written text where one exists. |
| **rule** | `string` | optional | The `name` of the ADVISORY validation rule this finding reports. Present only on an advisory rule hit. |
| **severity** | `Enum<'warning' \| 'info'>` | optional | The advisory rule's declared severity, `warning` or `info`. Present only on an advisory rule hit. |


---

## CreateManyDataRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **object** | `string` | ✅ | Object name |
| **records** | `Record<string, any>[]` | ✅ | Array of records to create |


---

## CreateManyDataResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **object** | `string` | ✅ | Object name |
| **records** | `Record<string, any>[]` | ✅ | Created records |
| **count** | `number` | ✅ | Number of records created |
| **droppedFields** | `{ object: string; fields: string[]; reason: Enum<'readonly' \| 'readonly_when' \| 'primary_key' \| 'computed' \| 'stored_expression'> }[]` | optional | Write-observability: caller-supplied `readonly` fields the in-engine create-side strip (`engine.insert`, `isSystem`-gated) removed before the rows were written. AGGREGATED across the batch (one event per object/reason with the UNION of dropped field names) rather than per-row, because this response is `{ object, records, count }` and has no per-row slot to hang a drop set on — a union is the only view it can represent. So read a name here as "at least one row dropped this field", NOT "every row dropped the same set": the strip runs INSIDE `engine.insert` after the `beforeInsert` hooks and exempts keys a hook itself wrote, tracked per row: rows where a hook stamped a protected key drop a different set from rows where it did not. Present ONLY when ≥1 field was dropped; the creates still succeeded without them (count/success unchanged). REST additionally surfaces this union as the `X-ObjectStack-Dropped-Fields` response header, each field and reason once. Optional — omit-when-empty keeps the shape backward-compatible. (The same reading applies to the partial-success bulk create, whose batch-level `droppedFields` names no row for the same reason. The paths that DO carry per-row `droppedFields` on each result are the bulk UPDATE and the mixed batch, where each row is its own engine call — see BatchOperationResultSchema.) |

### Nested Shape: `CreateManyDataResponse.droppedFields[number]`

A write-path strip event: caller-supplied fields legally dropped from the payload, or written with the engine's computed value in place of the caller's

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **object** | `string` | ✅ | Object the write targeted (resolved object name) |
| **fields** | `string[]` | ✅ | Caller-supplied field names whose value the engine did not store: removed from the write payload, or (`stored_expression`) written with the computed value in place of the caller's |
| **reason** | `Enum<'readonly' \| 'readonly_when' \| 'primary_key' \| 'computed' \| 'stored_expression'>` | ✅ | Why the fields were dropped: static readonly, a TRUE readonlyWhen predicate, the primary-key strip of a payload id the engine ruled is not an identifier, a computed (formula) field no driver has a column for, or a stored field whose declared expression the engine computes on every write, so the caller's different value was replaced by the computed one |


---

## DeleteDataRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **object** | `string` | ✅ | Object name |
| **id** | `string` | ✅ | Record ID to delete |
| **expectedVersion** | `string` | optional | Optimistic concurrency token (typically the `updated_at` value the client read). When provided, the server compares it against the current record version and returns 409 CONCURRENT_UPDATE if they differ. Optional — omit to skip the check. The quoted-empty entity-tag (`""`) is refused 400 VALIDATION_FAILED, not treated as omitted. |


---

## DeleteDataResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **object** | `string` | ✅ | Object name |
| **id** | `string` | ✅ | Deleted record ID |
| **success** | `boolean` | ✅ | Whether deletion succeeded |


---

## DeleteManyDataRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **ids** | `string[]` | ✅ | Array of record IDs to delete (server caps the count — see batch.maxBatchSize) |
| **options** | `{ atomic?: boolean; returnRecords?: boolean; continueOnError?: boolean }` | optional | Delete options |
| **object** | `string` | ✅ | Object name |

### Nested Shape: `DeleteManyDataRequest.options`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **atomic** | `boolean` | optional (default: `false`) | Opt-in all-or-nothing. When explicitly true the whole batch runs inside ONE engine transaction: the first failure rolls back every prior write, and the response reports zero successes — each row carries `errors[0].code` ROLLED_BACK (written, then undone), the causal row its own error, and rows never reached NOT_ATTEMPTED. A runtime that cannot roll back REFUSES the request (501 NOT_IMPLEMENTED) rather than silently degrading to best-effort — probe `capabilities.transactionalBatch` on /discovery first. Takes precedence over continueOnError. Default false: sequential best-effort. |
| **returnRecords** | `boolean` | optional (default: `false`) | If true, return full record data in response |
| **continueOnError** | `boolean` | optional (default: `false`) | If true (and atomic=false), continue processing remaining records after errors. Default false: the first failure ENDS the run — records before it stay written (nothing is rolled back on this arm), and every record after it is reported `errors[0].code` NOT_ATTEMPTED rather than omitted, so `results` always covers all `total` records and `succeeded + failed === total`. |
| **validateOnly** | `never` | optional | [REMOVED] `options.validateOnly` was removed from BatchOptions in @objectstack/spec 17.0.0. It was never implemented: the batch surfaces persisted regardless, so a "dry-run" would have silently executed. There is no dry-run today — drop the key. If you need to preview a batch without writing, open an issue so it can be designed (no-commit cascade / constraint semantics) and reintroduced as a flag that actually holds. |


---

## DeleteManyDataResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **success** | `boolean` | ✅ | Operation success status |
| **error** | `{ code: Enum<'VALIDATION_ERROR' \| 'INVALID_FIELD' \| 'MISSING_REQUIRED_FIELD' \| …>; declaredCode?: string; message: string; userMessage?: string; … }` | optional | Error details if success is false |
| **meta** | `{ timestamp: string; duration?: integer; requestId?: string; traceId?: string }` | optional | Response metadata |
| **operation** | `Enum<'create' \| 'update' \| 'upsert' \| 'delete'>` | optional | Operation type that was performed |
| **total** | `number` | ✅ | Total number of records in the batch |
| **succeeded** | `number` | ✅ | Number of records that succeeded |
| **failed** | `number` | ✅ | Number of records that failed |
| **results** | `{ id?: string; success: boolean; errors?: object[]; data?: Record<string, any>; … }[]` | ✅ | Detailed results for each record |

### Nested Shape: `DeleteManyDataResponse.error`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **code** | `Enum<'VALIDATION_ERROR' \| 'INVALID_FIELD' \| 'MISSING_REQUIRED_FIELD' \| …>` | ✅ | Error code (e.g. VALIDATION_ERROR; StandardErrorCode ∪ the ledger the serving side registers — ERROR_CODE_LEDGER for framework packages) |
| **declaredCode** | `string` | optional | The producer-declared code, verbatim, when it is not a member of the closed `code` vocabulary — the open, author-authored channel (app-specific spellings; ADR-0112) |
| **message** | `string` | ✅ | Readable error message |
| **userMessage** | `string` | optional | Producer-marked user-facing refusal text, verbatim. Present exactly when the producer opted in at throw time; consumers render it to end users and keep their generic substitution for anything unmarked. Status-agnostic; never replaces `message`. |
| **refusal** | `true` | optional | Producer-declared: the 5xx this envelope carries is a deliberate refusal whose `message` is authored for the caller, so a boundary that reads the declaration keeps it verbatim (until the withhold arms read it, a declared refusal is still withheld). Absent (the default) on a declared fault, whose `message` is withheld from the body and logged for the operator; redundant on a 4xx. Presence is the declaration — `true` is the only value. |
| **category** | `string` | optional | Error category (e.g. validation, authorization) |
| **httpStatus** | `integer` | optional | HTTP status of the response carrying this error |
| **details** | `any` | optional | Additional error context (e.g. field validation errors) |
| **requestId** | `string` | optional | Request ID for tracking |

### Nested Shape: `DeleteManyDataResponse.meta`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **timestamp** | `string` | ✅ |  |
| **duration** | `integer` | optional | Server-side processing duration in milliseconds |
| **requestId** | `string` | optional |  |
| **traceId** | `string` | optional |  |

### Nested Shape: `DeleteManyDataResponse.results[number]`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **id** | `string` | optional | Record ID if operation succeeded |
| **success** | `boolean` | ✅ | Whether this record was processed successfully |
| **errors** | `{ code: Enum<'VALIDATION_ERROR' \| 'INVALID_FIELD' \| 'MISSING_REQUIRED_FIELD' \| …>; declaredCode?: string; message: string; userMessage?: string; … }[]` | optional | Array of errors if operation failed. Branch on `errors[0].code` — an atomic batch that rolled back marks rows that were written then undone with code ROLLED_BACK and rows never reached with NOT_ATTEMPTED, while the causal row keeps its own error. A NON-atomic batch that stopped (the `continueOnError: false` default) marks its un-attempted tail with the same NOT_ATTEMPTED code — rows before the failure stay written and keep reporting success, since nothing was rolled back. |
| **data** | `Record<string, any>` | optional | Full record data (if returnRecords=true) |
| **index** | `number` | optional | Index of the record in the request array |
| **droppedFields** | `{ object: string; fields: string[]; reason: Enum<'readonly' \| 'readonly_when' \| 'primary_key' \| 'computed' \| 'stored_expression'> }[]` | optional | Write-observability: caller-supplied fields LEGALLY stripped from THIS row before it was written — static `readonly` / TRUE `readonlyWhen` on update, or the in-engine static `readonly` strip on create. Per-row because a batch can drop different fields on different rows (`readonlyWhen` is record-state-dependent). Present ONLY when ≥1 field was dropped for this row; the row still succeeded (success unchanged). A single response header cannot say which row dropped what, so this body field stays the canonical per-row channel; REST also sends the `X-ObjectStack-Dropped-Fields` response header on the answer, naming the union of the rows' drops (each field and reason once). Optional — omit-when-empty keeps the shape backward-compatible. |


---

## DeleteMetaItemRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **type** | `string` | ✅ | Metadata type name |
| **name** | `string` | ✅ | Item name |
| **parentVersion** | `string` | optional | ADR-0008 optimistic-concurrency pin: the version token the caller believes is current (on the REST door, the `If-Match` request header). Present, a concurrent edit is reported as a 409 conflict instead of silently reset; absent = last-write-wins against the current row (Studio's "Reset" button is unpinned). |
| **actor** | `string` | optional | Identity recorded on the delete's history tombstone row. On the REST door this is the request's authenticated identity (one producer) — never a caller-supplied header. Absent, the event is recorded actor-less (null), deliberately not attributed to "system". |
| **state** | `Enum<'active' \| 'draft'>` | optional | Which lifecycle row to discard: `draft` discards the pending draft overlay only (the still-active overlay, if any, keeps serving); `active` or absent resets the live row. Absent defaults to `active`. |
| **dropStorage** | `boolean` | optional | Destructive opt-in, default false: also drop the object's physical table after the metadata row is removed (`object` type + `active` state only; never `sys_` tables). Used by the "discard a previewed object" flow so a publish-to-preview leaves no orphan table. |
| **member** | `string` | optional | Bundle member selector, for a metadata type whose identity is a pair (today only `email_template`, identified by `(name, locale)`): the member this request addresses, spelled as its discriminator value (`zh-CN`). Absent, the canonical member (`en-US` for `email_template`), which is what a bare-name read answers; naming the canonical value is the same request. It is not the response translation locale, which `?locale=` and `Accept-Language` choose. Refused with 400 `VALIDATION_ERROR` on a type that has no members, and for a blank value or one longer than a stored member can be (100 characters). A well-formed member with nothing stored or shipped answers as that door answers an absent item. |
| **packageId** | `string \| null` | optional | ADR-0048 — the software package whose row this delete removes (`sys_metadata.package_id`; `?package=<id>` on the REST door, where `all` and an empty value name no package), as a save names the row it writes. Named, the delete removes that package's row and no other; a package that holds no row removes nothing. Absent or `null` names no package: the delete of the `active` row then removes the row of the one package the item's active rows are bound to, is refused `409 RESOURCE_CONFLICT` when they are bound to two or more, and with none removes the env-local row (no package stamp); the discard of the `draft` removes the draft a draft save naming no package writes, the one the per-item publish naming none promotes, and is refused where that publish is (see `PublishMetaItemRequestSchema.packageId`). That row is the one `parentVersion` is judged against. When another row of the item is still stored afterwards, the receipt `message` names it. |


---

## DeleteMetaItemResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **success** | `boolean` | ✅ |  |
| **reset** | `boolean` | optional |  |
| **seq** | `integer` | optional | Monotonic sequence number of the metadata event this delete appended to the item history (sys_metadata_history.event_seq) — the ordering token the history/audit trail is read by, and the same key both write-verb siblings declare. Unlike `version` it is not an OCC token. Optional HERE, unlike on those siblings, because only the repository path's delete-ful branch appends an event: a no-op reset (no overlay row existed) and the control-plane legacy raw-engine path both answer without one. Absence means "this branch appended no history event", never "nothing was deleted" — read `reset` for that. |
| **projectionApplied** | `{ success: boolean; error?: string }` | optional | Outcome of the awaited ADR-0094 mutation projector — the post-persist step that materializes this metadata into its derived data-plane read model. The same receipt `{@link SaveMetaItemResponseSchema}` and `{@link PublishMetaItemResponseSchema}` carry, because the projector runs on all three verbs of this door: on a delete it re-reads the layered state and either retires the derived record or resets it to the artifact baseline. Present ONLY when a projector is registered for this metadata type AND a row was actually deleted, which is why it is optional: its absence means "no projector ran", never "the projection failed". Best-effort by design — a projector failure is reported here and logged, never thrown, so a caller that needs the read model to be live must check `projectionApplied.success` rather than rely on the 200. |
| **message** | `string` | optional |  |

### Nested Shape: `DeleteMetaItemResponse.projectionApplied`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **success** | `boolean` | ✅ | False when the projector threw; the metadata delete itself still succeeded. |
| **error** | `string` | optional | Projector failure message, present only when `success` is false. |


---

## DiffMetaItemResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **type** | `string` | ✅ | Metadata type of the diffed item. |
| **name** | `string` | ✅ | Name of the diffed item. |
| **fromVersion** | `number \| null` | ✅ | The older side's history version, `null` when that side is absent (e.g. the item had no earlier version). |
| **toVersion** | `number \| null` | ✅ | The newer side's history version, `null` when that side is absent. |
| **added** | `{ path: string; value: any }[]` | ✅ | Members present in `to` and absent in `from`. |
| **removed** | `{ path: string; value: any }[]` | ✅ | Members present in `from` and absent in `to`. |
| **changed** | `{ path: string; from: any; to: any }[]` | ✅ | Members present on both sides with different values. |

### Nested Shape: `DiffMetaItemResponse.added[number]`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **path** | `string` | ✅ | Dot path of the added member. |
| **value** | `any` | ✅ | The added value. |

### Nested Shape: `DiffMetaItemResponse.removed[number]`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **path** | `string` | ✅ | Dot path of the removed member. |
| **value** | `any` | ✅ | The removed value. |

### Nested Shape: `DiffMetaItemResponse.changed[number]`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **path** | `string` | ✅ | Dot path of the changed member. |
| **from** | `any` | ✅ | The older side's value. |
| **to** | `any` | ✅ | The newer side's value. |


---

## DisablePackageRequest

Disable package request

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **id** | `string` | ✅ | Package ID to disable |


---

## DisablePackageResponse

Disable package response

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **package** | `{ manifest: object; source: Enum<'code' \| 'marketplace' \| 'authored'>; status?: Enum<'installed' \| 'disabled' \| 'installing' \| 'upgrading' \| 'uninstalling' \| 'error'>; enabled?: boolean; … }` | ✅ | Disabled package details |
| **message** | `string` | optional | Disable status message |

### Nested Shape: `DisablePackageResponse.package`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **manifest** | `{ id: string; namespace?: string; defaultDatasource?: string; version: string; … }` | ✅ | Package manifest at the AUTHORING stage; a row installed by a `defineStack()` host carries the assembled body instead — see `AssembledInstalledPackageSchema` / `InstalledPackageAtEitherStageSchema` |
| **source** | `Enum<'code' \| 'marketplace' \| 'authored'>` | ✅ | Where the installed package came from: code (registered by the deployment's own code), marketplace (installed from the marketplace), or authored (a database-authored package) |
| **status** | `Enum<'installed' \| 'disabled' \| 'installing' \| 'upgrading' \| 'uninstalling' \| 'error'>` | optional (default: `"installed"`) | Package state: installed, disabled, installing, upgrading, uninstalling, or error |
| **enabled** | `boolean` | optional (default: `true`) | Whether the package is currently enabled |
| **installedAt** | `string` | optional | Installation timestamp |
| **updatedAt** | `string` | optional | Last update timestamp |
| **installedVersion** | `string` | optional | Currently installed version for quick access |
| **previousVersion** | `string` | optional | Version before the last upgrade |
| **statusChangedAt** | `string` | optional | Status change timestamp |
| **errorMessage** | `string` | optional | Error message when status is error |
| **settings** | `Record<string, any>` | optional | User-provided configuration settings |
| **upgradeHistory** | `{ fromVersion: string; toVersion: string; upgradedAt: string; status: Enum<'success' \| 'failed' \| 'rolled_back'>; … }[]` | optional | Version upgrade history |
| **registeredNamespaces** | `string[]` | optional | Namespace prefixes registered by this package |


---

## EnablePackageRequest

Enable package request

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **id** | `string` | ✅ | Package ID to enable |


---

## EnablePackageResponse

Enable package response

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **package** | `{ manifest: object; source: Enum<'code' \| 'marketplace' \| 'authored'>; status?: Enum<'installed' \| 'disabled' \| 'installing' \| 'upgrading' \| 'uninstalling' \| 'error'>; enabled?: boolean; … }` | ✅ | Enabled package details |
| **message** | `string` | optional | Enable status message |

### Nested Shape: `EnablePackageResponse.package`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **manifest** | `{ id: string; namespace?: string; defaultDatasource?: string; version: string; … }` | ✅ | Package manifest at the AUTHORING stage; a row installed by a `defineStack()` host carries the assembled body instead — see `AssembledInstalledPackageSchema` / `InstalledPackageAtEitherStageSchema` |
| **source** | `Enum<'code' \| 'marketplace' \| 'authored'>` | ✅ | Where the installed package came from: code (registered by the deployment's own code), marketplace (installed from the marketplace), or authored (a database-authored package) |
| **status** | `Enum<'installed' \| 'disabled' \| 'installing' \| 'upgrading' \| 'uninstalling' \| 'error'>` | optional (default: `"installed"`) | Package state: installed, disabled, installing, upgrading, uninstalling, or error |
| **enabled** | `boolean` | optional (default: `true`) | Whether the package is currently enabled |
| **installedAt** | `string` | optional | Installation timestamp |
| **updatedAt** | `string` | optional | Last update timestamp |
| **installedVersion** | `string` | optional | Currently installed version for quick access |
| **previousVersion** | `string` | optional | Version before the last upgrade |
| **statusChangedAt** | `string` | optional | Status change timestamp |
| **errorMessage** | `string` | optional | Error message when status is error |
| **settings** | `Record<string, any>` | optional | User-provided configuration settings |
| **upgradeHistory** | `{ fromVersion: string; toVersion: string; upgradedAt: string; status: Enum<'success' \| 'failed' \| 'rolled_back'>; … }[]` | optional | Version upgrade history |
| **registeredNamespaces** | `string[]` | optional | Namespace prefixes registered by this package |


---

## FindDataRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **object** | `string` | ✅ | The unique machine name of the object to query (e.g. "account"). |
| **query** | `{ object: string; fields?: string[]; where?: [string, string, any] \| [string, string] \| [string, object, ...object[]] \| object[] \| Record<string, any> \| any; search?: string \| object; … }` | optional | Structured query definition (filter, sort, select, pagination) — the canonical QueryAST or its transport spelling. |

### Nested Shape: `FindDataRequest.query`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **object** | `string` | ✅ | Object name (e.g. account) |
| **fields** | `string[]` | optional | Fields to retrieve — names of the queried object's OWN columns, at least one: omit `fields` to read every column, because an empty list or a blank name is refused rather than read as every column. A dotted path (`owner.name`) is not a projection: no driver resolves one, and the ingress refuses it with `400 INVALID_FIELD`. Related data is read with `expand`, whose nested QueryAST both filters (`where`) and selects (`fields`) the related record's columns. The projection must RETAIN the foreign-key column: `fields: ['title']` with `expand: 'project_id'` resolves nothing, because the relation is carried by that key — add `'project_id'` and it works. Where the value is wanted on the queried object itself, denormalise it onto that object (a stored field, written when the source changes), the same remedy the sort axis prescribes. |
| **where** | `[string, string, any] \| [string, string] \| [string, object, ...object[]] \| object[] \| Record<string, any> \| any` | optional | Filtering criteria (WHERE) — a filter condition, or the input-only `FilterArray` sugar (`['status', '=', 'open']`), which is lowered through `parseFilterAST` before the query is produced. |
| **search** | `string \| { query: string; fields?: string[]; fuzzy?: boolean; operator?: Enum<'and' \| 'or'>; … }` | optional | Full-text search — the query text (canonical, ADR-0061 D1), or a structured FullTextSearch configuration |
| **searchFields** | `string[]` | optional | Narrow the search to these fields (server-intersected with the allowed searchable set — can only narrow, never widen; ADR-0061 D1) — at least one field name; a value naming no field (blank, only commas and spaces, or an empty list) is refused |
| **orderBy** | `{ field: string; order?: Enum<'asc' \| 'desc'> }[]` | optional | Sorting instructions (ORDER BY) |
| **limit** | `integer` | optional | Max records to return (LIMIT) |
| **offset** | `integer` | optional | Records to skip (OFFSET) |
| **top** | `integer` | optional | Alias for limit (OData compatibility) |
| **cursor** | `never` | optional | [REMOVED] `query.cursor` was removed in @objectstack/spec 17 (ADR-0049) — no driver ever implemented keyset pagination, so the cursor was accepted and ignored and every page came back identical (a caller looping "until hasMore is false" never terminates). Delete the key; `QueryBuilder.cursor()` was removed with it. Express the keyset as an ordinary `where` predicate on your sort key — `where: { created_at: { $gt: last.created_at } }` with the matching `orderBy` — which every driver executes with canonicalised comparands. A first-class cursor, if ever built, will be a response-minted opaque token, not this caller-built record. |
| **joins** | `never` | optional | [REMOVED] `query.joins` was removed in @objectstack/spec 17 (ADR-0049) — no engine or driver ever read it: a query carrying `joins` behaved exactly as if the key were absent, while its name squatted on the reserved REST parameter set. Delete the key. Related records are read through `expand` — `expand: { owner_id: { object: 'user', fields: ['name'] } }` — which the engine resolves via batch $in queries, and whose nested query selects the related record's own columns. Keep the foreign key in your own projection (`fields: ['title', 'owner_id']`): the relation is carried by that column, so projecting it away leaves expansion nothing to resolve. A dotted `fields` path is NOT a replacement — no driver ever resolved one and the ingress refuses it (`400 INVALID_FIELD`). |
| **aggregations** | `{ function: Enum<'count' \| 'sum' \| 'avg' \| 'min' \| 'max' \| 'count_distinct'>; field?: string; alias: string; filter?: any }[]` | optional | Aggregation functions |
| **groupBy** | `(string \| { field: string; dateGranularity?: Enum<'day' \| 'week' \| 'month' \| 'quarter' \| 'year'>; alias?: string })[]` | optional | GROUP BY targets (strings or `{field, dateGranularity?}` objects for date bucketing) |
| **having** | `any` | optional | HAVING — filter over the AGGREGATED rows (aggregation aliases + groupBy projections); applied engine-side after aggregation |
| **windowFunctions** | `never` | optional | [REMOVED] `query.windowFunctions` was removed in @objectstack/spec 17 (ADR-0049) — `find()` never applied it: no engine or driver read the key on the query path, so every OVER clause it declared was silently dropped. Delete the key. Window functions are a SQL-driver capability behind `SqlDriver.findWithWindowFunctions(object, query)` (embedder-level; not on the `IDataDriver` contract or the REST surface); request-level analytics are `aggregations` + `groupBy`. |
| **distinct** | `never` | optional | [REMOVED] `query.distinct` was removed in @objectstack/spec 17 (ADR-0049 / ADR-0078) — no driver ever rendered SELECT DISTINCT; the flag's only observable effect was MIS-WIRED: the REST list path treated a distinct query as not countable and silently degraded `total`/`hasMore` to a page-local estimate while still returning duplicate rows. Delete the key; `QueryBuilder.distinct()` was removed with it, and the count suppression is gone (`total` is truthful again). For unique values of one column use the SQL/memory drivers' `distinct(object, field)` door; for unique combinations, `groupBy`; for a deduplicated count, the `count_distinct` aggregation. |
| **expand** | `Record<string, { object: string; fields?: string[]; where?: any; search?: string \| object; … }>` | optional | Recursive relation loading map. Keys are lookup/master_detail field names; values are nested QueryAST objects that control select (`fields`) and filter (`where`, AND-merged with the batch $in), plus further expansion on the related object. The engine resolves expand via batch $in queries (driver-agnostic) with a default max depth of 3; per-parent `limit`/`offset`/`orderBy` are NOT applied on this path. |
| **$filter** | `[string, string, any] \| [string, string] \| [string, object, ...object[]] \| object[] \| Record<string, any> \| any \| null` | optional | Transport spelling of `where` (OData `$filter`) |
| **filters** | `[string, string, any] \| [string, string] \| [string, object, ...object[]] \| object[] \| Record<string, any> \| any \| null` | optional | Transport spelling of `where` (plural of `filter`) |
| **$top** | `integer \| string \| null` | optional | Transport spelling of `limit` (OData `$top`) — a non-negative integer, or the digits a querystring carries it as |
| **$skip** | `integer \| string \| null` | optional | Transport spelling of `offset` (OData `$skip`) — a non-negative integer, or the digits a querystring carries it as |
| **$orderby** | `Record<string, Enum<'asc' \| 'desc'>> \| Record<string, 1 \| -1> \| { field: string; order?: Enum<'asc' \| 'desc'> }[] \| null` | optional | Transport spelling of `orderBy` (OData `$orderby`) |
| **$select** | `string \| string[] \| null` | optional | Transport spelling of `fields` (OData `$select`) — at least one field name; a value naming no field (blank, only commas and spaces, an empty list, or a list with a blank name) is refused |
| **$expand** | `string \| string[] \| Record<string, { object: string; fields?: string[]; where?: any; search?: string \| object; … }> \| null` | optional | Transport spelling of `expand` (OData `$expand`) |
| **$search** | `string \| { query: string; fields?: string[]; fuzzy?: boolean; operator?: Enum<'and' \| 'or'>; … } \| null` | optional | Transport spelling of `search` (OData `$search`) |
| **$searchFields** | `string \| string[] \| null` | optional | Transport spelling of `searchFields` (OData `$searchFields`) — at least one field name; a value naming no field (blank, only commas and spaces, or an empty list) is refused |
| **$count** | `boolean \| Enum<'true' \| 'false'> \| null` | optional | Transport spelling of the response total-count flag (OData `$count`) |
| **count** | `boolean \| Enum<'true' \| 'false'> \| null` | optional | Response total-count flag — only an explicit `false` skips the COUNT query |
| **filter** | `[string, string, any] \| [string, string] \| [string, object, ...object[]] \| object[] \| Record<string, any> \| any \| null` | optional | Transport spelling of `where` |
| **select** | `string \| string[] \| null` | optional | Transport spelling of `fields` — at least one field name; a value naming no field (blank, only commas and spaces, an empty list, or a list with a blank name) is refused |
| **sort** | `Record<string, Enum<'asc' \| 'desc'>> \| Record<string, 1 \| -1> \| { field: string; order?: Enum<'asc' \| 'desc'> }[] \| null` | optional | Transport spelling of `orderBy` |
| **skip** | `integer \| string \| null` | optional | Transport spelling of `offset` — a non-negative integer, or the digits a querystring carries it as |
| **populate** | `string \| string[] \| null` | optional | Transport spelling of `expand` |


---

## FindDataResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **object** | `string` | ✅ | The object name for the returned records. |
| **records** | `Record<string, any>[]` | ✅ | The list of matching records. |
| **total** | `number` | optional | Total number of records matching the filter (if requested). |
| **nextCursor** | `string` | optional | Cursor for the next page of results (cursor-based pagination). |
| **hasMore** | `boolean` | optional | True if there are more records available (pagination). |


---

## FindReferencesToMetaResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **references** | `{ type: string; name: string; label?: string; path: string; … }[]` | ✅ | Every found reference to the addressed item. |

### Nested Shape: `FindReferencesToMetaResponse.references[number]`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **type** | `string` | ✅ | Metadata type of the REFERRING item. |
| **name** | `string` | ✅ | Name of the referring item. |
| **label** | `string` | optional | Display label of the referring item, when it has one. |
| **path** | `string` | ✅ | Where inside the referring item the reference sits (dot path). |
| **kind** | `string` | ✅ | What kind of reference this is (e.g. which key carries it). |


---

## FlowRegistration

### Union Options

This schema accepts one of the following structures:

#### Option 1

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **state** | `'registered'` | ✅ | The engine holds a registration under the flow's name. |

---

#### Option 2

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **state** | `'refused'` | ✅ | The engine refused to register the stored flow and holds nothing under its name: it never runs. |
| **reason** | `{ code?: string; message: string }` | ✅ | What the engine threw, verbatim. |

### Nested Shape: `FlowRegistration[state='refused'].reason`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **code** | `string` | optional | The machine code the engine's refusal carried, verbatim. Absent when the thrown error carried none. |
| **message** | `string` | ✅ | The engine's refusal message, verbatim. Render it as sent; do not parse it. |

---

#### Option 3

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **state** | `'unregistered'` | ✅ | The engine holds nothing under the flow's name and recorded no refusal for it. |

---


---

## FlowRegistrationRefusal

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **code** | `string` | optional | The machine code the engine's refusal carried, verbatim. Absent when the thrown error carried none. |
| **message** | `string` | ✅ | The engine's refusal message, verbatim. Render it as sent; do not parse it. |


---

## GetDataRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **object** | `string` | ✅ | The object name. |
| **id** | `string` | ✅ | The unique record identifier (primary key). |
| **select** | `string[]` | optional | Fields to include in the response (allowlisted query param) — at least one field name; an empty list, or a list with a blank name, is refused. |
| **expand** | `string[]` | optional | Lookup/master_detail field names to expand. The engine resolves these via batch $in queries, replacing foreign key IDs with full objects. |


---

## GetDataResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **object** | `string` | ✅ | The object name. |
| **id** | `string` | ✅ | The record ID. |
| **record** | `Record<string, any>` | ✅ | The complete record data. |


---

## GetDiscoveryRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |


---

## GetDiscoveryResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **name** | `string` | optional |  |
| **version** | `string` | ✅ |  |
| **environment** | `Enum<'production' \| 'sandbox' \| 'development'>` | optional | Deployment posture a discovery response advertises. Deliberately three coarse buckets — a client reads this to answer "am I talking to production?", not to identify a specific environment (that is `sys_environment` / EnvironmentTypeSchema, a richer 7-member taxonomy). |
| **routes** | `{ data: string; metadata: string; discovery?: string; ui?: string; … }` | optional |  |
| **locale** | `{ default: string; supported: string[]; timezone: string }` | optional |  |
| **services** | `Record<string, { enabled: boolean; status: Enum<'available' \| 'registered' \| 'unavailable' \| 'degraded' \| 'stub'>; handlerReady?: boolean; route?: string; … }>` | optional | Per-service availability map keyed by CoreServiceName |
| **capabilities** | `{ comments: object; automation: object; cron: object; search: object; … }` | optional | Hierarchical capability descriptors — the full WellKnownCapabilities vocabulary, every key present |
| **schemaDiscovery** | `{ openapi?: string; jsonSchema?: string }` | optional | Schema discovery endpoints for API toolchain integration |
| **scoping** | `{ enabled: boolean; resolution: Enum<'required' \| 'optional' \| 'auto'>; scoped: boolean; environmentId?: string }` | optional | Environment-scoping posture, added by the REST discovery endpoint |
| **authFamilies** | `{ admin: boolean }` | optional | Which optional better-auth route families are mounted under routes.auth — the same source as GET `{routes.auth}`/config features; absent when no auth service answers |
| **metadata** | `Record<string, any>` | optional | Custom metadata key-value pairs for extensibility |
| **apiName** | `string` | optional | API name (deprecated — use `name`; removed in protocol 18) |

### Nested Shape: `GetDiscoveryResponse.routes`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **data** | `string` | ✅ | e.g. /api/v1/data |
| **metadata** | `string` | ✅ | e.g. /api/v1/meta |
| **discovery** | `string` | optional | e.g. /api/v1/discovery |
| **ui** | `string` | optional | e.g. /api/v1/ui |
| **auth** | `string` | optional | e.g. /api/v1/auth |
| **automation** | `string` | optional | e.g. /api/v1/automation |
| **storage** | `string` | optional | e.g. /api/v1/storage |
| **analytics** | `string` | optional | e.g. /api/v1/analytics |
| **packages** | `string` | optional | e.g. /api/v1/packages |
| **datasources** | `string` | optional | e.g. /api/v1/datasources — base for the datasources/:name/external/* federation-admin family; absent when no host mounts it |
| **email** | `string` | optional | e.g. /api/v1/email — base for the email/send endpoint; absent when no host mounts it |
| **approvals** | `string` | optional | e.g. /api/v1/approvals |
| **realtime** | `string` | optional | e.g. /api/v1/realtime — present only when a realtime transport is actually mounted |
| **notifications** | `string` | optional | e.g. /api/v1/notifications |
| **ai** | `string` | optional | e.g. /api/v1/ai |
| **i18n** | `string` | optional | e.g. /api/v1/i18n |
| **mcp** | `string` | optional | e.g. /api/v1/mcp — always the unscoped base; absent when MCP is disabled or unserveable |

### Nested Shape: `GetDiscoveryResponse.services[string]`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **enabled** | `boolean` | ✅ | Whether the slot is filled by something this host delivers. NOT "a channel exists": subscribing requires handlerReady:true AND a connectable route (isSubscribableChannel). |
| **status** | `Enum<'available' \| 'registered' \| 'unavailable' \| 'degraded' \| 'stub'>` | ✅ | available = fully operational, registered = route declared but handler unverified, unavailable = not installed, degraded = partial, stub = placeholder that returns 501 |
| **handlerReady** | `boolean` | optional | Whether the HTTP handler is confirmed to be mounted. Omitted = readiness unknown/unverified; true = handler mounted; false = handler missing or stub (likely 501). |
| **route** | `string` | optional | e.g. /api/v1/analytics |
| **provider** | `string` | optional | e.g. "objectql", "plugin-redis", "driver-memory" |
| **version** | `string` | optional | Semantic version of the service implementation (e.g. "3.0.6") |
| **message** | `string` | optional | e.g. "Install plugin-workflow to enable" |
| **rateLimit** | `{ requestsPerMinute?: integer; requestsPerHour?: integer; burstLimit?: integer; retryAfterMs?: integer }` | optional | Rate limit and quota info for this service |

### Nested Shape: `GetDiscoveryResponse.capabilities`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **comments** | `{ enabled: boolean; features?: Record<string, boolean>; description?: string }` | ✅ | Whether the backend supports record comments / chatter (the `sys_comment` object served via the data API) |
| **automation** | `{ enabled: boolean; features?: Record<string, boolean>; description?: string }` | ✅ | Whether the backend supports Automation CRUD (flows, triggers) |
| **cron** | `{ enabled: boolean; features?: Record<string, boolean>; description?: string }` | ✅ | Whether the backend supports cron scheduling |
| **search** | `{ enabled: boolean; features?: Record<string, boolean>; description?: string }` | ✅ | Whether the backend supports full-text search |
| **export** | `{ enabled: boolean; features?: Record<string, boolean>; description?: string }` | ✅ | Whether the backend supports async export |
| **chunkedUpload** | `{ enabled: boolean; features?: Record<string, boolean>; description?: string }` | ✅ | Whether the backend supports chunked (multipart) uploads |
| **transactionalBatch** | `{ enabled: boolean; features?: Record<string, boolean>; description?: string }` | ✅ | Whether the backend exposes the atomic cross-object batch endpoint (POST `{basePath}`/batch, /ADR-0034): all ops commit or roll back together in one transaction. Lets clients skip non-atomic client-side simulation instead of runtime-probing 404/405/501. True ⟺ the /batch route is mounted AND the runtime can honour a transaction. |
| **websockets** | `{ enabled: boolean; features?: Record<string, boolean>; description?: string }` | ✅ | Whether the backend mounts a realtime push surface (WebSocket/SSE) clients can subscribe to. Derived from isSubscribableChannel(services.realtime): handlerReady true AND a connectable route. |
| **files** | `{ enabled: boolean; features?: Record<string, boolean>; description?: string }` | ✅ | Whether a file-storage surface (upload/download/attachments) is served |
| **analytics** | `{ enabled: boolean; features?: Record<string, boolean>; description?: string }` | ✅ | Whether the backend serves the analytics / BI query surface |
| **ai** | `{ enabled: boolean; features?: Record<string, boolean>; description?: string }` | ✅ | Whether the backend serves the AI surface (NLQ, chat, agents, suggest) |
| **notifications** | `{ enabled: boolean; features?: Record<string, boolean>; description?: string }` | ✅ | Whether the backend serves the notification surface (inbox, delivery) |
| **i18n** | `{ enabled: boolean; features?: Record<string, boolean>; description?: string }` | ✅ | Whether the backend serves the i18n surface (translations, locale negotiation) |

### Nested Shape: `GetDiscoveryResponse.schemaDiscovery`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **openapi** | `string` | optional | URL to OpenAPI (Swagger) specification (e.g., "/api/v1/openapi.json") |
| **jsonSchema** | `string` | optional | URL to JSON Schema definitions |

### Nested Shape: `GetDiscoveryResponse.scoping`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **enabled** | `boolean` | ✅ | Whether environment-scoped routes are mounted at all |
| **resolution** | `Enum<'required' \| 'optional' \| 'auto'>` | ✅ | How the environment id is resolved when scoping is enabled (mirrors RestApiConfig.projectResolution) |
| **scoped** | `boolean` | ✅ | Whether THIS response was served from the environment-scoped mount |
| **environmentId** | `string` | optional | The resolved environment id — present only on a scoped mount |

### Nested Shape: `GetDiscoveryResponse.authFamilies`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **admin** | `boolean` | ✅ | Whether the better-auth admin family (`{routes.auth}`/admin/*: list-users, set-role, update-user, ban-user, …) is mounted. Same value as features.admin on GET `{routes.auth}`/config, read from the same source; false means those routes answer a plain 404 on this deployment. |


---

## GetEffectivePermissionsRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |


---

## GetEffectivePermissionsResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **objects** | `Record<string, { allowCreate?: boolean; allowRead?: boolean; allowEdit?: boolean; allowDelete?: boolean; … }>` | ✅ | Effective object permissions keyed by object name |
| **systemPermissions** | `string[]` | ✅ | Effective system-level permissions |

### Nested Shape: `GetEffectivePermissionsResponse.objects[string]`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **allowCreate** | `boolean` | optional (default: `false`) | Create permission |
| **allowRead** | `boolean` | optional (default: `false`) | Read permission |
| **allowEdit** | `boolean` | optional (default: `false`) | Edit permission. Requires read on the same entry (allowRead, viewAllRecords or modifyAllRecords): an object permission that edits a record also reads it. |
| **allowDelete** | `boolean` | optional (default: `false`) | Delete permission. Requires read on the same entry (allowRead, viewAllRecords or modifyAllRecords): an object permission that deletes a record also reads it. |
| **allowExport** | `boolean` | optional | User-level export axis over read (opt-in grant). true = export granted (still bounded by read); unset/false = no export. Merged most-permissively like the CRUD bits; NOT implied by viewAllRecords/modifyAllRecords. |
| **allowTransfer** | `boolean` | optional (default: `false`) | [RBAC-gated; ENFORCED via the insert/update owner_id guard] Change record ownership (assign/reassign/disown owner_id) |
| **allowRestore** | `never` | optional | [REMOVED] `objects.<object>.allowRestore` was removed in @objectstack/spec 17 (ADR-0049) — the `restore` ObjectQL operation it claimed to gate has never shipped (roadmap M2), so granting the bit delivered nothing. Delete the key — a dispatched `restore` stays denied fail-closed by the permission evaluator's destructive-operation backstop, and the bit returns with the M2 lifecycle initiative alongside the operation it gates. Run `os migrate meta --from 17` to list the mechanical edits for existing sources; `--write` applies the ones it can prove, and you apply the rest by hand. |
| **allowPurge** | `never` | optional | [REMOVED] `objects.<object>.allowPurge` was removed in @objectstack/spec 17 (ADR-0049) — the `purge` ObjectQL operation it claimed to gate has never shipped (roadmap M2), so granting the bit delivered nothing (a compliance/GDPR erase the author believed was permission-locked was not — the operation itself does not exist). Delete the key — a dispatched `purge` stays denied fail-closed by the permission evaluator's destructive-operation backstop, and the bit returns with the M2 lifecycle initiative alongside the operation it gates. Run `os migrate meta --from 17` to list the mechanical edits for existing sources; `--write` applies the ones it can prove, and you apply the rest by hand. |
| **viewAllRecords** | `boolean` | optional (default: `false`) | View All Data (Bypass Sharing) |
| **modifyAllRecords** | `boolean` | optional (default: `false`) | Modify All Data (Bypass Sharing) — bypasses sharing rules and ownership on the objects record sharing enforces on; on an object with NO owner field sharing abstains, so the platform created_by write floor still applies. Includes delete: an omitted allowDelete parses to true, and an explicit allowDelete: false beside it is refused. |
| **readScope** | `Enum<'own' \| 'own_and_reports' \| 'unit' \| 'unit_and_below' \| 'org'>` | optional | [ADR-0057 D1] Read depth: own\|unit\|unit_and_below\|org |
| **writeScope** | `Enum<'own' \| 'own_and_reports' \| 'unit' \| 'unit_and_below' \| 'org'>` | optional | [ADR-0057 D1] Write depth: own\|unit\|unit_and_below\|org |
| **apiOperations** | `Enum<'get' \| 'list' \| 'create' \| 'update' \| 'delete' \| 'upsert' \| 'bulk' \| …>[]` | optional | Server-resolved effective API operations for this object. Present only when the object tightens exposure via apiMethods; absent = default-allow. The frontend renders this effective set, never the raw whitelist. Vocabulary is the EFFECTIVE ApiOperation set (six primitives + eight derived verbs), not the authored six-value ApiMethod enum. |


---

## GetFieldLabelsRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **object** | `string` | ✅ | Object name |
| **locale** | `string` | ✅ | BCP-47 locale code |


---

## GetFieldLabelsResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **object** | `string` | ✅ | Object name |
| **locale** | `string` | ✅ | Locale code |
| **labels** | `Record<string, { label: string; help?: string; options?: Record<string, string> }>` | ✅ | Field labels keyed by field name |

### Nested Shape: `GetFieldLabelsResponse.labels[string]`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **label** | `string` | ✅ | Translated field label |
| **help** | `string` | optional | Translated help text |
| **options** | `Record<string, string>` | optional | Translated option labels |


---

## GetLocalesRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |


---

## GetLocalesResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **locales** | `{ code: string; label: string; isDefault?: boolean }[]` | ✅ | Available locales |

### Nested Shape: `GetLocalesResponse.locales[number]`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **code** | `string` | ✅ | BCP-47 locale code (e.g., en-US, zh-CN) |
| **label** | `string` | ✅ | Locale label. Equals `code` on every serving surface today — the client names locales for its UI |
| **isDefault** | `boolean` | optional (default: `false`) | Whether this is the default locale |


---

## GetMetaDiagnosticsResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **entries** | `{ type: string; name: string; diagnostics: object }[]` | ✅ | One entry per item that failed validation (after filters). |
| **total** | `number` | ✅ | Number of entries in this answer. |
| **scannedTypes** | `number` | ✅ | How many metadata types the sweep visited. |
| **scannedItems** | `number` | ✅ | How many items the sweep visited. |
| **stats** | `Record<string, { count: number; locked: number; packages: string[] }>` | ✅ | Per-type aggregate stats, keyed by metadata type — computed in the same sweep so a directory page renders tile counts and a package filter in one round-trip. |

### Nested Shape: `GetMetaDiagnosticsResponse.entries[number]`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **type** | `string` | ✅ | Metadata type of the failing item. |
| **name** | `string` | ✅ | Name of the failing item. |
| **diagnostics** | `{ valid: boolean; errors?: object[]; warnings?: object[] }` | ✅ | The spec-validation verdict for the item — the same `MetadataValidationResult` the write path answers. |

### Nested Shape: `GetMetaDiagnosticsResponse.stats[string]`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **count** | `number` | ✅ | Items of this type present. |
| **locked** | `number` | ✅ | Items of this type currently lock-protected. |
| **packages** | `string[]` | ✅ | Packages contributing items of this type. |


---

## GetMetaItemCachedRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **type** | `string` | ✅ | Metadata type name |
| **name** | `string` | ✅ | Item name |
| **cacheRequest** | `{ ifNoneMatch?: string; ifModifiedSince?: string; cacheControl?: object }` | optional | Cache validation parameters |
| **locale** | `string` | optional | Resolved response locale. Folded into the ETag so a language switch never returns a stale-locale 304 — metadata is translated *after* the cache validator check (issue). |

### Nested Shape: `GetMetaItemCachedRequest.cacheRequest`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **ifNoneMatch** | `string` | optional | ETag value for conditional request (If-None-Match header) |
| **ifModifiedSince** | `string` | optional | Timestamp for conditional request (If-Modified-Since header) |
| **cacheControl** | `{ directives: Enum<'public' \| 'private' \| 'no-cache' \| 'no-store' \| 'must-revalidate' \| 'max-age'>[]; maxAge?: number; staleWhileRevalidate?: number; staleIfError?: number }` | optional | Client cache control preferences |


---

## GetMetaItemCachedResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **data** | `any` | optional | Metadata payload (omitted for 304 Not Modified) |
| **etag** | `{ value: string; weak?: boolean }` | optional | ETag for this resource version |
| **lastModified** | `string` | optional | Last modification timestamp |
| **cacheControl** | `{ directives: Enum<'public' \| 'private' \| 'no-cache' \| 'no-store' \| 'must-revalidate' \| 'max-age'>[]; maxAge?: number; staleWhileRevalidate?: number; staleIfError?: number }` | optional | Cache control directives |
| **notModified** | `boolean` | optional (default: `false`) | True if resource has not been modified (304 response) |
| **version** | `string` | optional | Metadata version identifier |
| **registration** | `{ state: 'registered' } \| { state: 'refused'; reason: object } \| { state: 'unregistered' }` | optional | The automation engine's verdict on the flow in `data`, read live on every call and never part of `etag` (see `FlowRegistrationSchema`). Present only for a `flow` read answered by a process that runs an automation service; a read that carries it is never answered `notModified`. Absent otherwise, and absence means no engine judged it, never `registered`. |
| **disabledActions** | `string[]` | optional | The names in the `actions` of the object in `data` that this deployment's activation ledger switches off (a call to any of them is refused 409 `ACTION_DISABLED`), read live on every call and never part of `etag`. Present only for an `object` read answered by a process whose data engine has an activation ledger attached, `[]` when none is switched off; this method never answers `notModified` for a read that carries it (the REST door folds the served list into the ETag it checks instead, so its 304 comes back only while the switched-off set is unchanged). Absent otherwise, and absence means no ledger answered, never that nothing is switched off. |
| **cascadeDeleteObjects** | `string[]` | optional | The objects, by name, that deleting a record of the object in `data` can also delete (the item read's `cascadeDeleteObjects`), read live on every call and never part of `etag`. Present only for an `object` read answered by a process whose data engine answers it, `[]` when nothing cascades; this method never answers `notModified` for a read that carries it (the REST door folds the served list into the ETag it checks instead). Absent otherwise, and absence means no engine answered, never that nothing cascades. |

### Nested Shape: `GetMetaItemCachedResponse.etag`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **value** | `string` | ✅ | ETag value (hash or version identifier) |
| **weak** | `boolean` | optional (default: `false`) | Whether this is a weak ETag |

### Nested Shape: `GetMetaItemCachedResponse.cacheControl`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **directives** | `Enum<'public' \| 'private' \| 'no-cache' \| 'no-store' \| 'must-revalidate' \| 'max-age'>[]` | ✅ | Cache control directives |
| **maxAge** | `number` | optional | Maximum cache age in seconds (unit per HTTP Cache-Control `max-age` (RFC 9111 §5.2.2.1)) |
| **staleWhileRevalidate** | `number` | optional | Allow serving stale content while revalidating (seconds) (unit per HTTP Cache-Control `stale-while-revalidate` (RFC 5861 §3)) |
| **staleIfError** | `number` | optional | Allow serving stale content on error (seconds) (unit per HTTP Cache-Control `stale-if-error` (RFC 5861 §4)) |

### Nested Shape: `GetMetaItemCachedResponse.registration[state='registered']`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **state** | `'registered'` | ✅ | The engine holds a registration under the flow's name. |

### Nested Shape: `GetMetaItemCachedResponse.registration[state='refused']`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **state** | `'refused'` | ✅ | The engine refused to register the stored flow and holds nothing under its name: it never runs. |
| **reason** | `{ code?: string; message: string }` | ✅ | What the engine threw, verbatim. |

### Nested Shape: `GetMetaItemCachedResponse.registration[state='unregistered']`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **state** | `'unregistered'` | ✅ | The engine holds nothing under the flow's name and recorded no refusal for it. |


---

## GetMetaItemLayeredRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **type** | `string` | ✅ | Metadata type name |
| **name** | `string` | ✅ | Item name |
| **packageId** | `string` | optional | Optional package ID — scopes the `code` layer so a same-name collision resolves to the requested package's artifact (ADR-0048), and the `draft` layer to that package's pending draft row. |
| **member** | `string` | optional | Bundle member selector, for a metadata type whose identity is a pair (today only `email_template`, identified by `(name, locale)`): the member this request addresses, spelled as its discriminator value (`zh-CN`). Absent, the canonical member (`en-US` for `email_template`), which is what a bare-name read answers; naming the canonical value is the same request. It is not the response translation locale, which `?locale=` and `Accept-Language` choose. Refused with 400 `VALIDATION_ERROR` on a type that has no members, and for a blank value or one longer than a stored member can be (100 characters). A well-formed member with nothing stored or shipped answers as that door answers an absent item. |


---

## GetMetaItemLayeredResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **type** | `string` | ✅ | Metadata type name (canonical singular) |
| **name** | `string` | ✅ | Item name |
| **code** | `any` | ✅ | LAYER 1 — the packaged artifact baseline exactly as shipped, before any tenant customization. `null` when no artifact ships this item (it exists only as an overlay). |
| **overlay** | `any` | ✅ | LAYER 2 — the stored customization row ALONE, not merged with `code`. `null` when the environment has not customized the item. |
| **overlayScope** | `Enum<'env'> \| null` | ✅ | Which scope the `overlay` row was read from: `env`, the environment layer — the only layer a read serves (ADR-0131 D6). `null` exactly when `overlay` is null. |
| **effective** | `any` | ✅ | LAYER 3 — the merged result, i.e. the value an ordinary `GET /meta/:type/:name` would return under `item`. `null` when the item resolves to nothing at all. |
| **draft** | `any` | optional | LAYER — the pending draft row exactly as stored: untranslated, ungoverned, with no injected columns (unlike the `?state=draft` item read, which serves the draft translated and governed). For the requested `packageId` when one is given. `null` when no draft is pending. A client building an edit buffer reads `draft ?? overlay ?? code`. ABSENT from a server that does not serve this layer: absence means "not served", never "no draft". |
| **_diagnostics** | `{ valid: boolean; errors?: object[]; warnings?: object[] }` | optional | Load-time spec-validation verdict for `effective`, so the Studio edit page can raise invalid-metadata banners and inline field errors without a second round trip. ABSENT for metadata types that register no Zod schema (function / service / router) — absence means "no opinion", never "valid". |
| **lock** | `Enum<'none' \| 'no-overlay' \| 'no-delete' \| 'full'>` | ✅ | Resolved lock verdict (ADR-0010 §3.3), artifact winning over overlay. Always present on this path. |
| **lockReason** | `string` | optional | Human-readable explanation shown next to a refused write. Present only when the resolved item declares `_lockReason`. |
| **lockSource** | `Enum<'artifact' \| 'package' \| 'env-forced'>` | optional | Which layer asserted the lock. Present only when the resolved item declares `_lockSource`. |
| **lockDocsUrl** | `string` | optional | Documentation link surfaced beside `lockReason`. Present only when the resolved item declares `_lockDocsUrl`. |
| **provenance** | `Enum<'package' \| 'org' \| 'env-forced'>` | optional | Where the item came from (package \| org \| env-forced). Present only when the resolved item declares `_provenance`. |
| **packageId** | `string` | optional | Owning package machine id. Present only when the resolved item declares `_packageId`. |
| **packageVersion** | `string` | optional | Owning package version. Present only when the resolved item declares `_packageVersion`. |
| **editable** | `boolean` | ✅ | Whether an overlay write is permitted. Always present on this path. |
| **deletable** | `boolean` | ✅ | Whether deleting the overlay is permitted. Always present on this path. |
| **resettable** | `boolean` | ✅ | Whether the item can be reset to its packaged default. Always present on this path. |

### Nested Shape: `GetMetaItemLayeredResponse._diagnostics`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **valid** | `boolean` | ✅ | Whether the metadata is valid |
| **errors** | `{ path: string; message: string; code?: string }[]` | optional | Validation errors |
| **warnings** | `{ path: string; message: string }[]` | optional | Validation warnings |


---

## GetMetaItemRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **type** | `string` | ✅ | Metadata type name |
| **name** | `string` | ✅ | Item name: the item's own `name`. For `type: 'app'` only, it may instead be the id of the package that ships the app (`manifest.id`, reverse-domain, e.g. `com.acme.crm`): with one app per package the app IS the package container, and the package id is its route key (ADR-0048 §3.1). The two grammars are disjoint (an app name is snake_case with no dot; a package id matches `MANIFEST_ID_PATTERN`, with at least one dot), so an app key in the package-id grammar always names the package (`isAppPackageIdKey`). That read answers the package's one app as the by-name read answers it, under the app's own `name`, and the per-caller read gate judges it as it judges that app read by name (an app withheld from the caller answers 403 `PERMISSION_DENIED`). A package with no app, or with more than one, answers as an absent item. A `packageId` naming a different package is refused with 400 `INVALID_REQUEST`. Every other type, and every other key, is read by name. |
| **packageId** | `string` | optional | Optional package ID to filter items by |
| **member** | `string` | optional | Bundle member selector, for a metadata type whose identity is a pair (today only `email_template`, identified by `(name, locale)`): the member this request addresses, spelled as its discriminator value (`zh-CN`). Absent, the canonical member (`en-US` for `email_template`), which is what a bare-name read answers; naming the canonical value is the same request. It is not the response translation locale, which `?locale=` and `Accept-Language` choose. Refused with 400 `VALIDATION_ERROR` on a type that has no members, and for a blank value or one longer than a stored member can be (100 characters). A well-formed member with nothing stored or shipped answers as that door answers an absent item. |
| **state** | `Enum<'active' \| 'draft'>` | optional | Draft-visibility switch — which lifecycle row to read (strict mode): `'draft'` opens the pending draft buffer (Studio's editor read) and fails when no draft exists; absent or `'active'` reads the live published row. Distinct from `previewDrafts`, which FALLS BACK to the active row when no draft exists. Declaration ≠ authorization: this member only selects which stored row is read — ADR-0106 masking is unaffected, and draft access is gated upstream, not by this schema. |
| **previewDrafts** | `boolean` | optional | Draft-visibility switch (ADR-0033 draft-overlay preview, non-strict): when true and `state` is not `'draft'`, a pending draft row is preferred if one exists, else the read falls back to the active row — the render path degrades to the published value instead of erroring. A served draft is tagged `_draft: true` so UIs can badge it. Declaration ≠ authorization: this member only switches which row is read, and ADR-0106 masking is unaffected — draft preview is admin-gated upstream, not by this schema. |


---

## GetMetaItemResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **type** | `string` | ✅ | Metadata type name |
| **name** | `string` | ✅ | Item name |
| **item** | `any` | ✅ | Metadata item definition |
| **sortability** | `{ fields: Record<string, object> }` | optional | Per-column sortability projection — present exactly when `type` is `object`, on every serving branch. Computed at serve time from the served document via the spec's own storage predicates; consumers render sort affordances from this signal and never re-derive it from field `type`. See `ObjectSortabilitySchema` for the closed category set. |
| **lock** | `Enum<'none' \| 'no-overlay' \| 'no-delete' \| 'full'>` | optional | Resolved lock verdict for this item (ADR-0010 §3.3). `none` means unlocked; `no-overlay` / `no-delete` / `full` mean the write doors refuse the corresponding write with 403. Joins two refusals: the document's own `_lock` (`ITEM_LOCKED`; the packaged artifact wins over any overlay), and the locked packaged base — an item a code package ships, on a type with no per-org overlay channel (`NOT_OVERRIDABLE`, or `ITEM_LOCKED` when the write names the read-only package). |
| **lockReason** | `string` | optional | Human-readable explanation shown next to a refused write. Present only when the resolved item declares `_lockReason`. |
| **lockSource** | `Enum<'artifact' \| 'package' \| 'env-forced'>` | optional | Which layer asserted the lock. Present only when the resolved item declares `_lockSource`. |
| **lockDocsUrl** | `string` | optional | Documentation link surfaced beside `lockReason`. Present only when the resolved item declares `_lockDocsUrl`. |
| **provenance** | `Enum<'package' \| 'org' \| 'env-forced'>` | optional | Where the item came from (package \| org \| env-forced). Present only when the resolved item declares `_provenance`. |
| **packageId** | `string` | optional | Owning package machine id. Present only when the resolved item declares `_packageId`. |
| **packageVersion** | `string` | optional | Owning package version. Present only when the resolved item declares `_packageVersion`. |
| **editable** | `boolean` | optional | Whether an overlay write is permitted — false iff `lock` is `no-overlay` or `full`. A derived verdict: do not recompute it from `lock` client-side. |
| **deletable** | `boolean` | optional | Whether deleting the overlay is permitted — false iff `lock` is `no-delete` or `full`. |
| **resettable** | `boolean` | optional | Whether the item can be reset to its packaged default — true iff it is artifact-backed, i.e. there is a baseline to reset TO. |
| **version** | `string \| null` | optional | ADR-0008 version token of the STORED row a save to this item compares against, at this read's scope (the environment's partition and `?package=`) and lifecycle (`?state=draft` reads the draft row, the plain read the active row). The same keyed token a save receipt's `version` serves for that row: send it back as the `If-Match` request header on `PUT /meta/:type/:name` (with `?mode=draft` for the draft) and a concurrent edit is refused 409 `METADATA_CONFLICT` instead of overwritten. `null` when no stored row is there — an item served from code or a package artifact, or from a row in another scope (a package-less row a package read falls back to) — so the next save creates one: pin it with `If-None-Match: *`. A `null` beside `editable: false` is no such invitation: that item takes no save at this address, so no pin is prescribed for it. ABSENT on the cached published-value branch and on a `?preview=draft` read, which publish no version: absence means "not published here", never "no row". Opaque — echo it verbatim, never parse it. |
| **registration** | `{ state: 'registered' } \| { state: 'refused'; reason: object } \| { state: 'unregistered' }` | optional | The automation engine's verdict on this flow (see `FlowRegistrationSchema`): `registered`, or `refused` with the engine's `reason` verbatim, or `unregistered`. Present only on a `flow` read answered by a process that runs an automation service; absent otherwise, and absence means no engine judged it, never `registered`. Unlike the protection keys it is published by the cached branch too, read live and never part of the ETag, so a cached read that carries it is never answered 304; it is absent on a draft read, whose pending body no engine holds. The verdict is the answering process's engine. Beside `item`, never inside it: the item's `status` stays what its author declared. |
| **disabledActions** | `string[]` | optional | The names in `item.actions` that this deployment's activation ledger switches off (an administrator turned them off under Setup › Packaged Actions): a call to any of them is refused 409 `ACTION_DISABLED`. Present only on an `object` read answered by a process whose data engine has an activation ledger attached, and then `[]` when none of this object's actions is switched off. Absent otherwise, and absence means no ledger answered, never that nothing is switched off. Unlike the protection keys it is published by the cached branch too, read live and folded into the ETag, so a cached read revalidates to 304 only while the switched-off set is unchanged; it is absent on a draft read. The ledger is the answering process's engine's. Beside `item`, never inside it: `item.actions` is served as declared, a switched-off action included. |
| **cascadeDeleteObjects** | `string[]` | optional | The objects, by name, that deleting a record of this object can also delete: every object whose relation into it cascades (a `master_detail` by default, a `lookup`, `tree` or `user` field that declares `deleteBehavior: 'cascade'`), and the objects that cascade into those in turn, each once, in the order the delete reaches them. This object itself is named only when a cascading relation leads back to it. A `restrict` or `set_null` relation removes nothing and is not named. Read from the rule the delete runs, and object-level: it never says which records or how many. Present only on an `object` read answered by a process whose data engine answers it, and then `[]` when nothing cascades. Absent otherwise, and absence means no engine answered, never that nothing cascades. Unlike the protection keys it is published by the cached branch too, read live and folded into the ETag; it is absent on a draft read. Beside `item`, never inside it: it is derived from other objects' definitions and is not authorable. |
| **draftChanges** | `{ path: string[]; by: Enum<'copilot' \| 'mcp' \| 'studio' \| 'api'>; actor: string \| null; client?: object; … }[]` | optional | Per-key provenance of the pending draft: which authoring surface last changed each key path of `item`, under which actor, and when. The draft save records it: each draft save is compared with the body that save replaced (the previous draft, or the stored published row for the draft's first save; a draft of an item with no stored published row is recorded as `[]`, the whole body). A `migrate-stored` rewrite of a draft row records nothing, since it is not authoring. The entries are kept from the first recorded save until the draft is published or discarded, which remove the draft row and the entries with it; and no two entries share a path: a save replaces every entry at or under each path it changes, so the last write to change a key is the entry whose `path` is the longest prefix of that key's path, and entries under a key's path are changes inside it. The entries name which writes changed what; they are not a list of every difference from the published body (compare the two bodies for that). Served only on a `?state=draft` read that answers a stored draft row. ABSENT on the active read, the cached branch and a `?preview=draft` read, and for a draft row whose saves recorded none: absence means "not recorded", never "no changes". |

### Nested Shape: `GetMetaItemResponse.sortability`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **fields** | `Record<string, { sortable: boolean; reason?: Enum<'virtual-type' \| 'author-declared'>; caveat?: 'unprovisioned-anchor' }>` | ✅ | Verdict per sortable-addressable column, keyed by field name. The domain is the served field map plus the always-provisioned `id`, less any field the caller's own field permission withholds: a caller exempt from the object-schema mask (ADR-0106 D4) is served such fields in `item` and gets no entry for them; for every other caller the served map already excludes them. A name absent from this map (an unknown field, a dotted path, an unprovisioned audit column, a field the caller may not read) has no platform sort behind it and must get no sort affordance. |

### Nested Shape: `GetMetaItemResponse.registration[state='registered']`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **state** | `'registered'` | ✅ | The engine holds a registration under the flow's name. |

### Nested Shape: `GetMetaItemResponse.registration[state='refused']`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **state** | `'refused'` | ✅ | The engine refused to register the stored flow and holds nothing under its name: it never runs. |
| **reason** | `{ code?: string; message: string }` | ✅ | What the engine threw, verbatim. |

### Nested Shape: `GetMetaItemResponse.registration[state='unregistered']`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **state** | `'unregistered'` | ✅ | The engine holds nothing under the flow's name and recorded no refusal for it. |

### Nested Shape: `GetMetaItemResponse.draftChanges[number]`

One changed key path of the pending draft, and the last write that changed it.

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **path** | `string[]` | ✅ | The changed key path inside `item`: one object key per segment, from the body root down to the deepest key whose value the write changed, added or removed. `["fields", "renewal_count"]` is a field the write added; `["fields", "renewal_count", "label"]` is one property of it. A segment is the key exactly as stored, dots included (a permission set's `fields` map is keyed `object.field`), so a path is never one dotted string. An array is one value: its path ends at its key (`["columns"]`) and never carries an index, which an insert or a reorder would shift. Which elements of an array the draft added or dropped is read by comparing it with the published body's array; the entry answers only who last changed the array as a whole. `[]` is the whole body: the write that created the draft of an item with no published body. |
| **by** | `Enum<'copilot' \| 'mcp' \| 'studio' \| 'api'>` | ✅ | The authoring surface of the write that last changed this path: the `surface` that save declared (`SaveMetaItemRequestSchema.surface`), `api` when it declared none. Closed: `copilot` is an AI assistant's tool writing on a person's behalf (Studio's AI copilot); `mcp` is an AI agent outside the console writing on a person's behalf through the MCP build door; `studio` is a person editing in Studio; `api` is any other caller of the save door. It names the surface, not the person: a copilot write, an MCP write and a Studio write of the same user carry the same `actor`. |
| **actor** | `string \| null` | ✅ | The identity that write was recorded under: the save's `actor`, the value its history event carries. `null` when the write recorded none, never a sentinel string. |
| **client** | `{ id: string; name?: string }` | optional | The OAuth client that write was made by: the save's `client`, which only the MCP build door states, so it is present only beside `by: 'mcp'`. ABSENT on every other surface, and on an `mcp` entry whose save stated none: absence means "not recorded", never "no client". |
| **at** | `string` | ✅ | When that write happened: an ISO-8601 instant in UTC, written with `Z` (the form `toISOString()` produces). An offset or a local time is refused. |


---

## GetMetaItemsRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **type** | `string` | ✅ | Metadata type name (e.g., "object", "plugin") |
| **packageId** | `string` | optional | Optional package ID to filter items by |
| **names** | `string[]` | optional | [PLANNED — not yet served] Item names to narrow the list to: only the named items of the type are answered, among those the list would answer without it. Sent over HTTP as one comma-separated value (`?names=a,b,c`). A name that matches no item is simply absent from the answer (no 404). An empty value is refused with 400 rather than read as every item; leave the key out to list every item. No list door reads it yet: until one does, a request carrying it is answered as if it were absent. |
| **previewDrafts** | `boolean` | optional | Draft-visibility switch (ADR-0033 draft-overlay preview): when true, pending `state='draft'` rows are overlaid on the active list — draft wins on name collision, draft-only items appear, and each overlaid item is tagged `_draft: true` so UIs can badge the preview. Absent/false = published world only. Declaration ≠ authorization: this member only switches which rows are read, and ADR-0106 masking is unaffected — callers without draft-preview authorization are refused upstream (admin-gated), not by this schema. |


---

## GetMetaItemsResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **type** | `string` | ✅ | Metadata type name |
| **items** | `any[]` | ✅ | Array of metadata items |
| **registrations** | `Record<string, { state: 'registered' } \| { state: 'refused'; reason: object } \| { state: 'unregistered' }>` | optional | The automation engine's verdict on each listed flow, keyed by the item's `name`: one entry per listed item that is not a pending draft. Present only on a `flow` list answered by a process that runs an automation service; absent otherwise, and absence means no engine judged them, never `registered`. The verdict is the answering process's engine. Beside the items, never inside one: an item's `status` stays what its author declared. |
| **disabledActions** | `Record<string, string[]>` | optional | For each listed object, keyed by the item's `name`: the names in its `actions` that this deployment's activation ledger switches off, so a call to any of them is refused 409 `ACTION_DISABLED` (the item read's `disabledActions`). One entry per listed object that is not a pending draft, `[]` when none of its actions is switched off. Present only on an `object` list answered by a process whose data engine has an activation ledger attached; absent otherwise, and absence means no ledger answered, never that nothing is switched off. Beside the items, never inside one: each item's `actions` is served as declared. |

### Nested Shape: `GetMetaItemsResponse.registrations[string][state='registered']`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **state** | `'registered'` | ✅ | The engine holds a registration under the flow's name. |

### Nested Shape: `GetMetaItemsResponse.registrations[string][state='refused']`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **state** | `'refused'` | ✅ | The engine refused to register the stored flow and holds nothing under its name: it never runs. |
| **reason** | `{ code?: string; message: string }` | ✅ | What the engine threw, verbatim. |

### Nested Shape: `GetMetaItemsResponse.registrations[string][state='unregistered']`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **state** | `'unregistered'` | ✅ | The engine holds nothing under the flow's name and recorded no refusal for it. |


---

## GetMetaTypesRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |


---

## GetMetaTypesResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **types** | `string[]` | ✅ | Available metadata type names (e.g., "object", "plugin", "view") |
| **entries** | `{ type: string; label: string; description?: string; filePatterns: string[]; … }[]` | optional | Enriched per-type registry entries (Phase 3a) |

### Nested Shape: `GetMetaTypesResponse.entries[number]`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **type** | `string` | ✅ | Singular type identifier |
| **label** | `string` | ✅ | Human-readable label |
| **description** | `string` | optional | Brief description |
| **filePatterns** | `string[]` | ✅ | Glob patterns used to discover artifacts of this type |
| **supportsOverlay** | `boolean` | ✅ | Whether a stored overlay row of an item a managed package ships may be removed through the ordinary delete door without the type's environment-overlay permission: the removal removes no managed definition, and changes what is served for permission alone; it opens no save and it decides no read |
| **allowEnvironmentOverlay** | `boolean` | ✅ | True when the registry grants environment overlays of this type's packaged items, or when OS_METADATA_WRITABLE names the type (overrideSource env), which opens the type's writes of items no managed package ships and never a managed item (ADR-0131 D6) |
| **allowRuntimeCreate** | `boolean` | ✅ | New artifacts of this type can be created via runtime API |
| **supportsVersioning** | `boolean` | ✅ | History is tracked for this type |
| **executionPinned** | `boolean` | ✅ | Runtime transactions pin a specific historical version_hash (ADR-0009) |
| **loadOrder** | `integer` | ✅ | Loading priority (lower = earlier) |
| **domain** | `Enum<'data' \| 'ui' \| 'automation' \| 'system' \| 'security' \| 'ai'>` | ✅ | Protocol domain |
| **overrideSource** | `Enum<'registry' \| 'env'>` | ✅ | Whether allowEnvironmentOverlay comes from the static registry or is elevated by the OS_METADATA_WRITABLE environment variable |
| **createSeed** | `any` | optional | Authoritative minimal valid create seed for this type — Studio/CLI/API derive create defaults from it (single source of truth in @objectstack/spec). Absent for canvas-create types whose shape is built interactively. |


---

## GetNotificationPreferencesRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |


---

## GetNotificationPreferencesResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **preferences** | `{ email?: boolean; push?: boolean; inApp?: boolean; digest?: Enum<'none' \| 'daily' \| 'weekly'>; … }` | ✅ | Current notification preferences |

### Nested Shape: `GetNotificationPreferencesResponse.preferences`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **email** | `boolean` | optional (default: `true`) | Receive email notifications |
| **push** | `boolean` | optional (default: `true`) | Receive push notifications |
| **inApp** | `boolean` | optional (default: `true`) | Receive in-app notifications |
| **digest** | `Enum<'none' \| 'daily' \| 'weekly'>` | optional (default: `"none"`) | Email digest frequency |
| **channels** | `Record<string, { enabled?: boolean; email?: boolean; push?: boolean }>` | optional | Per-channel notification preferences |


---

## GetObjectPermissionsRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **object** | `string` | ✅ | Object name to get permissions for |


---

## GetObjectPermissionsResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **object** | `string` | ✅ | Object name |
| **permissions** | `{ allowCreate?: boolean; allowRead?: boolean; allowEdit?: boolean; allowDelete?: boolean; … }` | ✅ | Object-level permissions |
| **fieldPermissions** | `Record<string, { readable?: boolean; editable?: boolean; when?: string }>` | optional | Field-level permissions keyed by field name |

### Nested Shape: `GetObjectPermissionsResponse.permissions`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **allowCreate** | `boolean` | optional (default: `false`) | Create permission |
| **allowRead** | `boolean` | optional (default: `false`) | Read permission |
| **allowEdit** | `boolean` | optional (default: `false`) | Edit permission. Requires read on the same entry (allowRead, viewAllRecords or modifyAllRecords): an object permission that edits a record also reads it. |
| **allowDelete** | `boolean` | optional (default: `false`) | Delete permission. Requires read on the same entry (allowRead, viewAllRecords or modifyAllRecords): an object permission that deletes a record also reads it. |
| **allowExport** | `boolean` | optional | User-level export axis over read (opt-in grant). true = export granted (still bounded by read); unset/false = no export. Merged most-permissively like the CRUD bits; NOT implied by viewAllRecords/modifyAllRecords. |
| **allowTransfer** | `boolean` | optional (default: `false`) | [RBAC-gated; ENFORCED via the insert/update owner_id guard] Change record ownership (assign/reassign/disown owner_id) |
| **allowRestore** | `never` | optional | [REMOVED] `objects.<object>.allowRestore` was removed in @objectstack/spec 17 (ADR-0049) — the `restore` ObjectQL operation it claimed to gate has never shipped (roadmap M2), so granting the bit delivered nothing. Delete the key — a dispatched `restore` stays denied fail-closed by the permission evaluator's destructive-operation backstop, and the bit returns with the M2 lifecycle initiative alongside the operation it gates. Run `os migrate meta --from 17` to list the mechanical edits for existing sources; `--write` applies the ones it can prove, and you apply the rest by hand. |
| **allowPurge** | `never` | optional | [REMOVED] `objects.<object>.allowPurge` was removed in @objectstack/spec 17 (ADR-0049) — the `purge` ObjectQL operation it claimed to gate has never shipped (roadmap M2), so granting the bit delivered nothing (a compliance/GDPR erase the author believed was permission-locked was not — the operation itself does not exist). Delete the key — a dispatched `purge` stays denied fail-closed by the permission evaluator's destructive-operation backstop, and the bit returns with the M2 lifecycle initiative alongside the operation it gates. Run `os migrate meta --from 17` to list the mechanical edits for existing sources; `--write` applies the ones it can prove, and you apply the rest by hand. |
| **viewAllRecords** | `boolean` | optional (default: `false`) | View All Data (Bypass Sharing) |
| **modifyAllRecords** | `boolean` | optional (default: `false`) | Modify All Data (Bypass Sharing) — bypasses sharing rules and ownership on the objects record sharing enforces on; on an object with NO owner field sharing abstains, so the platform created_by write floor still applies. Includes delete: an omitted allowDelete parses to true, and an explicit allowDelete: false beside it is refused. |
| **readScope** | `Enum<'own' \| 'own_and_reports' \| 'unit' \| 'unit_and_below' \| 'org'>` | optional | [ADR-0057 D1] Read depth: own\|unit\|unit_and_below\|org |
| **writeScope** | `Enum<'own' \| 'own_and_reports' \| 'unit' \| 'unit_and_below' \| 'org'>` | optional | [ADR-0057 D1] Write depth: own\|unit\|unit_and_below\|org |

### Nested Shape: `GetObjectPermissionsResponse.fieldPermissions[string]`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **readable** | `boolean` | optional (default: `true`) | Field read access |
| **editable** | `boolean` | optional (default: `false`) | Field edit access |
| **when** | `string` | optional | Row condition narrowing where this entry votes, authored in canonical CEL like `rowLevelSecurity[].using` (a field of this object compared with a literal or a `current_user.*` value using `==`, `!=`, `<`, `<=`, `>`, `>=` or `in`, combined with `&&` / `\|\|`; e.g. status in ["approved", "signing", "active"]). Absent, the entry votes on every row. Present, the entry votes as declared on the rows the condition matches and is ABSENT on every other row, so the most-permissive merge across the caller's sets is unchanged: a set that locks a field only from a stage onward does not lock the rows before it, and a co-held set that says nothing about the field still lifts nothing, unless it holds the super-user bypass on the object (its own entry for the object, otherwise its `'*'` wildcard). Such a set votes on every field of the object, on every row, whatever `when` says: `viewAllRecords` votes `readable: true`, and `modifyAllRecords` votes `readable: true` and `editable: true`, so a row-conditioned lock never binds a Modify All holder. View All never grants an edit. Otherwise union without deny is unchanged. The write gate judges the condition on the stored row BEFORE an update (so a lock on `status in [...]` holds on an approved row even when the write changes `status`) and on the incoming row of an insert, as the caller sent it. It reaches the EDIT vote only — reads are judged row-blind — so `readable: false` beside a `when` is refused at parse time: keep `readable: true` (the default) on a row-conditioned entry. Where no row can be judged (a predicate update with no row address, a stored row the caller cannot read, a condition that does not resolve for the request) the entry is judged as its lock: it contributes no edit grant on that write. Validated like an RLS predicate at authoring and publish time: the fields it names must exist on the entry's object and it may not reach across a relation; a condition that cannot be lowered is refused there, never dropped at runtime. Readers that hold no row (`/auth/me/permissions`, the object-wide field projections, the write preview on an update) report a row-conditioned entry as its lock, so a client never advertises an edit the gate refuses. The record-scoped field projection (`GET /data/{object}/{id}/field-access`, the answer a record's edit form reads) holds the row: it judges the condition on the stored row exactly as the write gate does on an update of that row, so a lock that applies only from a stage onward disables the field only on the rows it names. |


---

## GetPackageRequest

Get package request

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **id** | `string` | ✅ | Package identifier |


---

## GetPackageResponse

Get package response

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **package** | `{ manifest: object; source: Enum<'code' \| 'marketplace' \| 'authored'>; status?: Enum<'installed' \| 'disabled' \| 'installing' \| 'upgrading' \| 'uninstalling' \| 'error'>; enabled?: boolean; … }` | ✅ | Package details |

### Nested Shape: `GetPackageResponse.package`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **manifest** | `{ id: string; namespace?: string; defaultDatasource?: string; version: string; … }` | ✅ | Package manifest at the AUTHORING stage; a row installed by a `defineStack()` host carries the assembled body instead — see `AssembledInstalledPackageSchema` / `InstalledPackageAtEitherStageSchema` |
| **source** | `Enum<'code' \| 'marketplace' \| 'authored'>` | ✅ | Where the installed package came from: code (registered by the deployment's own code), marketplace (installed from the marketplace), or authored (a database-authored package) |
| **status** | `Enum<'installed' \| 'disabled' \| 'installing' \| 'upgrading' \| 'uninstalling' \| 'error'>` | optional (default: `"installed"`) | Package state: installed, disabled, installing, upgrading, uninstalling, or error |
| **enabled** | `boolean` | optional (default: `true`) | Whether the package is currently enabled |
| **installedAt** | `string` | optional | Installation timestamp |
| **updatedAt** | `string` | optional | Last update timestamp |
| **installedVersion** | `string` | optional | Currently installed version for quick access |
| **previousVersion** | `string` | optional | Version before the last upgrade |
| **statusChangedAt** | `string` | optional | Status change timestamp |
| **errorMessage** | `string` | optional | Error message when status is error |
| **settings** | `Record<string, any>` | optional | User-provided configuration settings |
| **upgradeHistory** | `{ fromVersion: string; toVersion: string; upgradedAt: string; status: Enum<'success' \| 'failed' \| 'rolled_back'>; … }[]` | optional | Version upgrade history |
| **registeredNamespaces** | `string[]` | optional | Namespace prefixes registered by this package |


---

## GetPresenceRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **channel** | `string` | ✅ | Channel to get presence for |


---

## GetPresenceResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **channel** | `string` | ✅ | Channel name |
| **members** | `{ userId: string; status: Enum<'online' \| 'away' \| 'busy' \| 'offline'>; lastSeen: string; metadata?: Record<string, any> }[]` | ✅ | Active members and their presence state |

### Nested Shape: `GetPresenceResponse.members[number]`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **userId** | `string` | ✅ | User identifier |
| **status** | `Enum<'online' \| 'away' \| 'busy' \| 'offline'>` | ✅ | Current presence status |
| **lastSeen** | `string` | ✅ | ISO 8601 datetime of last activity |
| **metadata** | `Record<string, any>` | optional | Custom presence data (e.g., current page, custom status) |


---

## GetPublishedMetaItemResponse

The published metadata item body, opaque by ruling (1C). Shape is the item's own metadata-type schema, resolved at read time — never frozen into this contract.


---

## GetRecordFieldAccessResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **object** | `string` | ✅ | Object name |
| **recordId** | `string` | ✅ | The record the projection was judged on |
| **readable** | `string[]` | ✅ | Fields the caller may read on this object. Reads are judged row-blind, so this equals the object-wide read projection |
| **editable** | `string[]` | ✅ | Fields the caller may write on this record, as field-level security judges an update of the stored row: a row-conditioned field permission locks a field only on the rows its condition names. A field absent here is refused by the write gate. The object grant, the record's row-level write scope (sharing, row-level security, ownership; see `writable`) and a field's own rules (readonly, computed types) are not part of it |
| **writable** | `boolean` | ✅ | Whether the caller may update this record at all, judged by the write gate on the stored row: the object's edit permission, record sharing and ownership, row-level security, and a parent record's edit access for a record controlled by its parent. False means an update that leaves the record as it is stored is refused: either the caller may not edit this record, or the stored record fails a row-level security check that every save must satisfy, which a save moving the checked value back inside that check can still pass. True does not promise that a save succeeds: a pending approval that locks the record, conditional read-only and validation rules on the values sent, a data source that refuses writes, the platform's own rules for its own records (comments, attachments, mail messages, approval requests) and an object the API serves read-only are judged when the record is saved |


---

## GetTranslationsRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **locale** | `string` | ✅ | BCP-47 locale code |


---

## GetTranslationsResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **locale** | `string` | ✅ | Locale code |
| **requestedLocale** | `string` | optional | The locale the caller asked for, when the served locale is a fallback from it |
| **translations** | `{ objects?: Record<string, object>; picklists?: Record<string, object>; apps?: Record<string, object>; messages?: Record<string, string>; … }` | ✅ | Translation data |

### Nested Shape: `GetTranslationsResponse.translations`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **objects** | `Record<string, { label?: string; pluralLabel?: string; description?: string; fields?: Record<string, object>; … }>` | optional | Object translations keyed by object name |
| **picklists** | `Record<string, { label?: string; options: Record<string, string> }>` | optional | Picklist translations keyed by picklist name |
| **apps** | `Record<string, { label?: string; description?: string; navigation?: Record<string, object> }>` | optional | App translations keyed by app name |
| **messages** | `Record<string, string>` | optional | UI message translations keyed by message ID |
| **globalActions** | `Record<string, { label?: string; description?: string; confirmText?: string; successMessage?: string; … }>` | optional | Global action translations keyed by action name |
| **dashboards** | `Record<string, { label?: string; description?: string; actions?: Record<string, object>; widgets?: Record<string, object>; … }>` | optional | Dashboard translations keyed by dashboard name |
| **datasets** | `Record<string, { label?: string; description?: string; dimensions?: Record<string, object>; measures?: Record<string, object> }>` | optional | Analytics dataset translations keyed by dataset name |
| **reports** | `Record<string, { label?: string; description?: string; blocks?: Record<string, object> }>` | optional | Report translations keyed by report name |
| **pages** | `Record<string, { label?: string; description?: string; title?: string; subtitle?: string; … }>` | optional | Page translations keyed by page name |
| **flows** | `Record<string, { label?: string; successMessage?: string; errorMessage?: string; screens?: Record<string, object>; … }>` | optional | Screen-flow translations keyed by flow name |
| **hooks** | `Record<string, { refusals?: Record<string, object> }>` | optional | Hook translations keyed by hook name |
| **positions** | `Record<string, { label?: string; description?: string }>` | optional | Position translations keyed by position name |
| **permissions** | `Record<string, { label?: string; description?: string }>` | optional | Permission-set translations keyed by permission-set name |
| **books** | `Record<string, { label?: string; description?: string; groups?: Record<string, object> }>` | optional | Doc-book translations keyed by book name |
| **settingsCommon** | `{ sourceLabels?: object }` | optional | Cross-namespace Settings UI strings |
| **settings** | `Record<string, { title?: string; description?: string; groups?: Record<string, object>; keys?: Record<string, object>; … }>` | optional | Settings manifest translations keyed by namespace |


---

## GetUiViewRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **object** | `string` | ✅ | Object name (snake_case) |
| **type** | `Enum<'list' \| 'form'>` | ✅ | View type |


---

## GetUiViewResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **name** | `string` | optional | Item name — supplied by the metadata door; for an object-scoped container it is the object name. |
| **label** | `string \| Record<string, string>` | optional | Human-readable label shown in metadata lists. |
| **object** | `string` | optional | Object this container binds to — how a stack-level `views: [...]` entry says which object its views belong to; read by `getViewsByObject()` / `GET /meta/view?object=`. |
| **list** | `{ name?: string; label?: string \| Record<string, string>; type?: Enum<'grid' \| 'kanban' \| 'gallery' \| 'calendar' \| 'timeline' \| 'gantt' \| 'map' \| 'chart' \| 'tree'>; data?: object \| object \| object; … }` | optional |  |
| **form** | `{ type?: Enum<'simple' \| 'tabbed' \| 'wizard'>; layout?: Enum<'vertical' \| 'horizontal'>; columns?: integer; title?: string; … }` | optional |  |
| **listViews** | `Record<string, { name?: string; label?: string \| Record<string, string>; type?: Enum<'grid' \| 'kanban' \| 'gallery' \| 'calendar' \| 'timeline' \| 'gantt' \| 'map' \| 'chart' \| 'tree'>; data?: object \| object \| object; … }>` | optional | Additional named list views (views mode — dropdown userFilters allowed, no tabs; ADR-0047) |
| **formViews** | `Record<string, { type?: Enum<'simple' \| 'tabbed' \| 'wizard'>; layout?: Enum<'vertical' \| 'horizontal'>; columns?: integer; title?: string; … }>` | optional | Additional named form views |
| **protection** | `{ lock: Enum<'none' \| 'no-overlay' \| 'no-delete' \| 'full'>; reason: string; docsUrl?: string }` | optional | Package author protection block — lock policy for this view. |
| **_lock** | `Enum<'none' \| 'no-overlay' \| 'no-delete' \| 'full'>` | optional | Item-level lock — controls overlay & delete (ADR-0010). |
| **_lockReason** | `string` | optional | Human-readable reason shown when a write is refused by _lock. |
| **_lockSource** | `Enum<'artifact' \| 'package' \| 'env-forced'>` | optional | Layer that set _lock (artifact \| package \| env-forced). |
| **_provenance** | `Enum<'package' \| 'org' \| 'env-forced'>` | optional | Origin of the item (package \| org \| env-forced). |
| **_packageId** | `string` | optional | Owning package machine id. |
| **_packageVersion** | `string` | optional | Owning package version. |
| **_lockDocsUrl** | `string` | optional | Optional documentation link surfaced next to _lockReason. |

### Nested Shape: `GetUiViewResponse.list`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **name** | `string` | optional | Internal view name (lowercase snake_case) |
| **label** | `string \| Record<string, string>` | optional | Display label — the default-language string, or an inline locale map (`{ en, "zh-CN" }`) resolved at render time |
| **type** | `Enum<'grid' \| 'kanban' \| 'gallery' \| 'calendar' \| 'timeline' \| 'gantt' \| 'map' \| 'chart' \| 'tree'>` | optional (default: `"grid"`) |  |
| **data** | `{ provider: 'object'; object: string } \| { provider: 'value'; items: any[] } \| { provider: 'schema'; schemaId: string; schema?: Record<string, any> }` | optional | Data source configuration (defaults to "object" provider) |
| **columns** | `string[] \| { field: string; label?: string \| Record<string, string>; width?: number; align?: Enum<'left' \| 'center' \| 'right'>; … }[]` | ✅ | Fields to display as columns — the PROJECTION of the per-view field composition `columns` x `hiddenFields` x `fieldOrder`: this list is the candidate set AND the baseline order; `hiddenFields` subtracts from it and `fieldOrder` orders what survives. `hiddenFields` and `fieldOrder` cannot add a field omitted here. An empty list declares no projection, so neither of them applies: which columns show is then left to the renderer (objectui's `ListView` grid derives the object's default columns). |
| **filter** | `({ field: string; operator: Enum<'equals' \| 'not_equals' \| 'contains' \| 'not_contains' \| 'icontains' \| …>; value?: string \| number \| boolean \| null \| (string \| number)[] } \| { or: object[] })[]` | optional | Filter criteria (JSON Rules) |
| **sort** | `{ field: string; order: Enum<'asc' \| 'desc'> }[]` | optional |  |
| **searchableFields** | `string[]` | optional | Fields enabled for search |
| **filterableFields** | `never` | optional | [REMOVED] `view.list.filterableFields` was removed in @objectstack/spec 18 (ADR-0049 enforce-or-remove) — the key was declared as a shorthand for `userFilters.fields` (the quick-filter bar), but the one renderer that read it used it as a whitelist of the fields the Filter panel offers instead, so one key carried two meanings and drew no quick filter. Delete the key; no per-view Filter-panel whitelist replaces it. For a curated end-user filter bar, declare `userFilters: { element: 'dropdown', fields: [{ field: '<field_name>' }] }`. The Filter panel offers every field the user may read: field read security decides, and a field the object definition marks `hidden: true` is not offered. Run `os migrate meta --from 17` to list the mechanical edits for existing sources; `--write` applies the ones it can prove, and you apply the rest by hand. |
| **resizable** | `boolean` | optional | Enable column resizing |
| **compactToolbar** | `boolean` | optional | Collapse Group/Color/Density/Hide-fields into a single View settings popover |
| **selection** | `{ type?: Enum<'none' \| 'single' \| 'multiple'> }` | optional | Row selection configuration |
| **navigation** | `{ mode?: Enum<'drawer' \| 'page' \| 'new_window' \| 'none'>; preventNavigation?: boolean; openNewTab?: boolean; size?: Enum<'auto' \| 'sm' \| 'md' \| 'lg' \| 'xl' \| 'full'>; … }` | optional | Configuration for item click navigation (drawer by default, page, new_window or none) |
| **pagination** | `{ pageSize?: integer; pageSizeOptions?: integer[] }` | optional | Pagination configuration |
| **kanban** | `{ groupByField: string; lanes?: string[]; summarizeField?: string; titleField?: string; … }` | optional | Kanban-board configuration — applies when the view renders as a kanban layout |
| **calendar** | `{ startDateField: string; endDateField?: string; titleField?: string; colorField?: string; … }` | optional | Calendar configuration — applies when the view renders as a calendar layout |
| **gantt** | `{ startDateField: string; endDateField: string; titleField: string; progressField?: string; … }` | optional | Gantt-timeline configuration — applies when the view renders as a gantt layout |
| **gallery** | `{ coverField?: string; coverFit?: Enum<'cover' \| 'contain'>; cardSize?: Enum<'small' \| 'medium' \| 'large'>; titleField?: string; … }` | optional | Gallery/card view configuration |
| **timeline** | `{ startDateField: string; endDateField?: string; titleField: string; groupByField?: string; … }` | optional | Timeline view configuration |
| **chart** | `{ chartType?: Enum<'bar' \| 'line' \| 'pie' \| 'area' \| 'scatter'>; dataset: string; dimensions?: string[]; values: string[] }` | optional | Chart binding — applies when the view renders as a chart. A `type: 'chart'` view must bind one: it names the ADR-0021 `dataset` and the measures (`values`) the chart plots, and there is no default binding |
| **map** | `{ latitudeField?: string; longitudeField?: string; locationField?: string; titleField?: string; … }` | optional | Map configuration — applies when the view renders as a map layout |
| **tree** | `{ parentField?: string; labelField?: string; fields?: string[]; defaultExpandedDepth?: integer }` | optional | Tree/hierarchy configuration — applies when the view renders as a tree layout |
| **pageName** | `never` | optional | [REMOVED] `view.pageName` was removed in @objectstack/spec 17.5.0 (ADR-0049 enforce-or-remove) — it named the page a `type: 'page'` view was to mount, and that mount was never built: no renderer read the key, so the named page was never reached and the view drew an empty grid. Delete the key; to put a published page in front of users, give the app a navigation item — `{ type: 'page', pageName: '<page_name>' }` under the app's `navigation` — which is a different key on a different surface and is the page mount that has always rendered. Run `os migrate meta --from 17` to list the mechanical edits for existing sources; `--write` applies the ones it can prove, and you apply the rest by hand. |
| **description** | `string \| Record<string, string>` | optional | View description for documentation/tooltips |
| **sharing** | `{ lockedBy?: string }` | optional | View sharing configuration: who locked the view configuration |
| **requiredPermissions** | `string[]` | optional | [ADR-0066] Capabilities a user must ALL hold for this list view or dashboard to be served to them — names that permission sets grant through `systemPermissions`, the same list as `requiredPermissions` on an app, a navigation item or an action. Absent or empty: no audience gate. There is no any-of form: to serve one item to several audiences (legal OR admin), declare one capability and grant it to each of their permission sets. The server enforces it: a user who does not hold them all is not listed the item and is refused it by name; a view container's list views are pruned the same way, and so are an object's own list views for a user who may not edit the object. |
| **rowHeight** | `Enum<'compact' \| 'short' \| 'medium' \| 'tall' \| 'extra_tall'>` | optional | Row height / density setting |
| **grouping** | `{ fields: object[] }` | optional | Group records by one or more fields — server-side: the groups and their header numbers come from an aggregate query over the whole filtered set, rows within a group are paged (see GroupingConfigSchema) |
| **rowColor** | `{ field: string; colors?: Record<string, string> }` | optional | Color rows based on field value |
| **hiddenFields** | `string[]` | optional | Fields to hide in this specific view — the SUBTRACTION of the per-view field composition `columns` x `hiddenFields` x `fieldOrder`: each name listed here is removed from the `columns` projection before `fieldOrder` orders the remainder. A name `columns` never projected subtracts nothing. |
| **fieldOrder** | `string[]` | optional | Explicit field display order for this view — the ORDERING of the per-view field composition `columns` x `hiddenFields` x `fieldOrder`: it sorts what survives `columns` minus `hiddenFields` and never adds a field. A surviving column absent from `fieldOrder` sorts LAST, after every listed one, keeping its `columns`-relative order; a name listed here that did not survive orders nothing. |
| **rowActions** | `string[]` | optional | Actions available for individual row items |
| **bulkActions** | `string[]` | optional | Actions available when multiple rows are selected |
| **bulkActionDefs** | `{ name: string; label?: string; icon?: string; variant?: Enum<'primary' \| 'secondary' \| 'danger' \| 'ghost' \| 'outline'>; … }[]` | optional | Rich bulk action definitions (schema-driven, executed via BulkActionDialog). Use a def for a mass data-plane mutation ('update' with a `patch` / 'delete') that no action expresses, or for an `operation: 'custom'` + `execution: 'aggregate'` entry that dispatches the action it NAMES once for the whole selection — the renderer injects `params._selectedIds: string[]` (read that on the server, not `recordId`) so a single call can produce one aggregate artifact (zip of QR codes, merged PDF, batch print). Aggregate results are all-or-nothing: a handler that cannot cover the whole selection must reject, and per-row retry is replaced by re-running the action. `batchSize` does not apply (the call is never chunked); set `maxRecords` on defs whose server work is expensive. For the PER-RECORD dispatch use `bulkActions: ['<name>']` instead — the bare-string form, promoted with the action's own label, params and `visible`; a 'custom' def without `execution: 'aggregate'` has no dispatcher and is refused at parse time. Toolbar url/api actions can also interpolate the current selection via `${ctx.selection.ids}` / `${ctx.selection.count}`. |
| **conditionalFormatting** | `{ condition: string \| object; style: Record<string, string> }[]` | optional | Conditional formatting rules for list rows |
| **inlineEdit** | `boolean` | optional | Allow inline editing of records directly in the list view |
| **exportOptions** | `Enum<'csv' \| 'xlsx' \| 'json'>[] \| { formats?: Enum<'csv' \| 'xlsx' \| 'json'>[]; maxRecords?: integer; includeHeaders?: boolean; fileNamePrefix?: string; … }` | optional | Export configuration for the list toolbar export menu: `{ formats?, maxRecords?, includeHeaders?, fileNamePrefix?, streaming? }`. A bare format array is the legacy spelling and lifts to `{ formats: [...] }` at parse. |
| **userActions** | `{ sort?: boolean; search?: boolean; filter?: boolean; refresh?: boolean; … }` | optional | User action toggles for the view toolbar |
| **appearance** | `{ showDescription?: boolean; allowedVisualizations?: Enum<'grid' \| 'kanban' \| 'gallery' \| 'calendar' \| 'timeline' \| 'gantt' \| 'map' \| 'chart' \| 'tree'>[] }` | optional | Appearance and visualization configuration |
| **tabs** | `never` | optional | [REMOVED] `view.list.tabs` was removed in @objectstack/spec 17.5.0 (ADR-0049 enforce-or-remove) — no renderer ever mounted a tab bar for it, so authoring it drew nothing: the tab strip above an object's records is the saved-view switcher (ViewTabBar), which renders one tab per named list view and never read this key. Delete the key, and move each tab you want to a named list view under the object's `listViews` instead: the tab's `name` becomes the entry's key, its `label` the entry's `label`, and its `filter` rules join the view's own `filter` on that entry (copy the view's `columns` too); a tab whose `view` already named a list view needs nothing more. Every `listViews` entry renders as a tab in the switcher. Run `os migrate meta --from 17` to list the mechanical edits for existing sources; `--write` applies the ones it can prove, and you apply the rest by hand. |
| **addRecord** | `{ enabled: boolean; position?: Enum<'top' \| 'bottom' \| 'both'> }` | optional | Add record entry point configuration |
| **showRecordCount** | `boolean` | optional | Show record count at the bottom of the list |
| **allowPrinting** | `boolean` | optional | Allow users to print the view |
| **emptyState** | `{ title?: string \| Record<string, string>; message?: string \| Record<string, string>; icon?: string; link?: object }` | optional | Empty state configuration when no records found |
| **aria** | `{ ariaLabel?: string \| Record<string, string>; ariaDescribedBy?: string; role?: string }` | optional | ARIA accessibility attributes for the list view |
| **userFilters** | `{ element?: Enum<'dropdown' \| 'toggle'>; fields?: object[] }` | optional |  |

### Nested Shape: `GetUiViewResponse.form`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **type** | `Enum<'simple' \| 'tabbed' \| 'wizard'>` | optional (default: `"simple"`) | Form layout: 'simple' (stacked sections, the default), 'tabbed' (one tab per section) or 'wizard' (one gated step per section). Where the form opens is the caller's choice, not the form's |
| **layout** | `Enum<'vertical' \| 'horizontal'>` | optional | Field layout direction — 'vertical' (the renderer default) or 'horizontal'. Multi-column is not a layout value: set `columns` |
| **columns** | `integer` | optional | Number of columns for the form body |
| **title** | `string` | optional | Form title |
| **description** | `string` | optional | Form description |
| **defaultTab** | `string` | optional | Initially active tab (tabbed forms) |
| **tabPosition** | `Enum<'top' \| 'bottom' \| 'left' \| 'right'>` | optional | Tab strip position (tabbed forms) |
| **allowSkip** | `boolean` | optional | Wizard step-gate opt-out: allow entering a later step without submitting the one before it. Navigation freedom, NOT a validation exemption — the final submit still re-validates every step and returns to the first failing one. Default (absent/false): steps unlock in array order as each prior step submits validly. UI admission only, never authorization. |
| **showStepIndicator** | `boolean` | optional | Show the wizard step indicator (renderer default: shown). Step progress (completed/current/upcoming/invalid) is derived from the step gate — this boolean is the only authorable progress surface. |
| **splitDirection** | `never` | optional | [REMOVED] `view.form.splitDirection` was removed in @objectstack/spec 18.3.0 (ADR-0050) — it set the pane orientation of a split form, and the split form renderer retired in the same release, so no form draws panes. Delete the key. Run `os migrate meta --from 17` to list the mechanical edits for existing sources; `--write` applies the ones it can prove, and you apply the rest by hand. |
| **splitSize** | `never` | optional | [REMOVED] `view.form.splitSize` was removed in @objectstack/spec 18.3.0 (ADR-0050) — it sized the primary pane of a `type: 'split'` form, and `split` left the form view `type` in the same release, so no form view draws panes. Delete the key. Run `os migrate meta --from 17` to list the mechanical edits for existing sources; `--write` applies the ones it can prove, and you apply the rest by hand. |
| **splitResizable** | `never` | optional | [REMOVED] `view.form.splitResizable` was removed in @objectstack/spec 18.3.0 (ADR-0050) — it made the panes of a `type: 'split'` form resizable, and `split` left the form view `type` in the same release, so no form view draws panes. Delete the key. Run `os migrate meta --from 17` to list the mechanical edits for existing sources; `--write` applies the ones it can prove, and you apply the rest by hand. |
| **drawerSide** | `never` | optional | [REMOVED] `view.form.drawerSide` was removed in @objectstack/spec 18.3.0 (ADR-0049) — no form view opens in a drawer, so no form view ever read it. Delete the key: a list opens a clicked record in its drawer (`navigation.mode`), and the React `ObjectForm` block keeps its own `drawerSide` prop for the drawer it opens with `formType='drawer'`. Run `os migrate meta --from 17` to list the mechanical edits for existing sources; `--write` applies the ones it can prove, and you apply the rest by hand. |
| **drawerWidth** | `never` | optional | [REMOVED] `view.form.drawerWidth` was removed in @objectstack/spec 18.3.0 (ADR-0049) — no form view opens in a drawer, so no form view ever read it, and a pixel width cannot be chosen without knowing the client viewport. Delete the key: a list's record drawer takes its starting width from the `navigation.size` bucket, and the React `ObjectForm` block keeps its own `drawerWidth` prop for the drawer it opens. Run `os migrate meta --from 17` to list the mechanical edits for existing sources; `--write` applies the ones it can prove, and you apply the rest by hand. |
| **modalSize** | `Enum<'sm' \| 'default' \| 'lg' \| 'xl' \| 'full'>` | optional | Size bucket of the dialog a caller opens this form in: the console's New / Edit dialog, or the modal container of the React `ObjectForm` block (formType 'modal') |
| **data** | `{ provider: 'object'; object: string } \| { provider: 'value'; items: any[] } \| { provider: 'schema'; schemaId: string; schema?: Record<string, any> }` | optional | Data source configuration (defaults to "object" provider) |
| **sections** | `{ name?: string; label?: string \| Record<string, string>; description?: string; collapsible?: boolean; … }[]` | optional |  |
| **groups** | `{ name?: string; label?: string \| Record<string, string>; description?: string; collapsible?: boolean; … }[]` | optional | [LEGACY ALIAS → `sections`] Accepted for back-compat and folded onto `sections` at parse; `sections` wins when both are present. Prefer `sections`. |
| **subforms** | `{ childObject: string; relationshipField?: string; columns?: object[]; amountField?: string; … }[]` | optional | Inline master-detail child collections |
| **sharing** | `{ enabled?: boolean; publicLink?: string; password?: string; allowedDomains?: string[]; … }` | optional | Public sharing configuration for this form |
| **submitBehavior** | `{ kind: 'thank-you'; title?: string; message?: string } \| { kind: 'redirect'; url: string; delayMs?: integer } \| { kind: 'continue' } \| { kind: 'next-record' }` | optional | Post-submit behavior. On the `redirect` arm, `url` is relative-only and interpolates only declared record fields as `{{record.field_name}}`, URL-escaped. |
| **buttons** | `{ submit?: object; cancel?: object }` | optional | Form action-button visibility & labels; folded onto the flat renderer props by ObjectUI ObjectForm. |
| **defaults** | `Record<string, any>` | optional | Initial field values for create-mode forms (folded into ObjectUI ObjectForm initial values;). |

### Nested Shape: `GetUiViewResponse.listViews[string]`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **name** | `string` | optional | Internal view name (lowercase snake_case) |
| **label** | `string \| Record<string, string>` | optional | Display label — the default-language string, or an inline locale map (`{ en, "zh-CN" }`) resolved at render time |
| **type** | `Enum<'grid' \| 'kanban' \| 'gallery' \| 'calendar' \| 'timeline' \| 'gantt' \| 'map' \| 'chart' \| 'tree'>` | optional (default: `"grid"`) |  |
| **data** | `{ provider: 'object'; object: string } \| { provider: 'value'; items: any[] } \| { provider: 'schema'; schemaId: string; schema?: Record<string, any> }` | optional | Data source configuration (defaults to "object" provider) |
| **columns** | `string[] \| { field: string; label?: string \| Record<string, string>; width?: number; align?: Enum<'left' \| 'center' \| 'right'>; … }[]` | ✅ | Fields to display as columns — the PROJECTION of the per-view field composition `columns` x `hiddenFields` x `fieldOrder`: this list is the candidate set AND the baseline order; `hiddenFields` subtracts from it and `fieldOrder` orders what survives. `hiddenFields` and `fieldOrder` cannot add a field omitted here. An empty list declares no projection, so neither of them applies: which columns show is then left to the renderer (objectui's `ListView` grid derives the object's default columns). |
| **filter** | `({ field: string; operator: Enum<'equals' \| 'not_equals' \| 'contains' \| 'not_contains' \| 'icontains' \| …>; value?: string \| number \| boolean \| null \| (string \| number)[] } \| { or: object[] })[]` | optional | Filter criteria (JSON Rules) |
| **sort** | `{ field: string; order: Enum<'asc' \| 'desc'> }[]` | optional |  |
| **searchableFields** | `string[]` | optional | Fields enabled for search |
| **filterableFields** | `never` | optional | [REMOVED] `view.list.filterableFields` was removed in @objectstack/spec 18 (ADR-0049 enforce-or-remove) — the key was declared as a shorthand for `userFilters.fields` (the quick-filter bar), but the one renderer that read it used it as a whitelist of the fields the Filter panel offers instead, so one key carried two meanings and drew no quick filter. Delete the key; no per-view Filter-panel whitelist replaces it. For a curated end-user filter bar, declare `userFilters: { element: 'dropdown', fields: [{ field: '<field_name>' }] }`. The Filter panel offers every field the user may read: field read security decides, and a field the object definition marks `hidden: true` is not offered. Run `os migrate meta --from 17` to list the mechanical edits for existing sources; `--write` applies the ones it can prove, and you apply the rest by hand. |
| **resizable** | `boolean` | optional | Enable column resizing |
| **compactToolbar** | `boolean` | optional | Collapse Group/Color/Density/Hide-fields into a single View settings popover |
| **selection** | `{ type?: Enum<'none' \| 'single' \| 'multiple'> }` | optional | Row selection configuration |
| **navigation** | `{ mode?: Enum<'drawer' \| 'page' \| 'new_window' \| 'none'>; preventNavigation?: boolean; openNewTab?: boolean; size?: Enum<'auto' \| 'sm' \| 'md' \| 'lg' \| 'xl' \| 'full'>; … }` | optional | Configuration for item click navigation (drawer by default, page, new_window or none) |
| **pagination** | `{ pageSize?: integer; pageSizeOptions?: integer[] }` | optional | Pagination configuration |
| **kanban** | `{ groupByField: string; lanes?: string[]; summarizeField?: string; titleField?: string; … }` | optional | Kanban-board configuration — applies when the view renders as a kanban layout |
| **calendar** | `{ startDateField: string; endDateField?: string; titleField?: string; colorField?: string; … }` | optional | Calendar configuration — applies when the view renders as a calendar layout |
| **gantt** | `{ startDateField: string; endDateField: string; titleField: string; progressField?: string; … }` | optional | Gantt-timeline configuration — applies when the view renders as a gantt layout |
| **gallery** | `{ coverField?: string; coverFit?: Enum<'cover' \| 'contain'>; cardSize?: Enum<'small' \| 'medium' \| 'large'>; titleField?: string; … }` | optional | Gallery/card view configuration |
| **timeline** | `{ startDateField: string; endDateField?: string; titleField: string; groupByField?: string; … }` | optional | Timeline view configuration |
| **chart** | `{ chartType?: Enum<'bar' \| 'line' \| 'pie' \| 'area' \| 'scatter'>; dataset: string; dimensions?: string[]; values: string[] }` | optional | Chart binding — applies when the view renders as a chart. A `type: 'chart'` view must bind one: it names the ADR-0021 `dataset` and the measures (`values`) the chart plots, and there is no default binding |
| **map** | `{ latitudeField?: string; longitudeField?: string; locationField?: string; titleField?: string; … }` | optional | Map configuration — applies when the view renders as a map layout |
| **tree** | `{ parentField?: string; labelField?: string; fields?: string[]; defaultExpandedDepth?: integer }` | optional | Tree/hierarchy configuration — applies when the view renders as a tree layout |
| **pageName** | `never` | optional | [REMOVED] `view.pageName` was removed in @objectstack/spec 17.5.0 (ADR-0049 enforce-or-remove) — it named the page a `type: 'page'` view was to mount, and that mount was never built: no renderer read the key, so the named page was never reached and the view drew an empty grid. Delete the key; to put a published page in front of users, give the app a navigation item — `{ type: 'page', pageName: '<page_name>' }` under the app's `navigation` — which is a different key on a different surface and is the page mount that has always rendered. Run `os migrate meta --from 17` to list the mechanical edits for existing sources; `--write` applies the ones it can prove, and you apply the rest by hand. |
| **description** | `string \| Record<string, string>` | optional | View description for documentation/tooltips |
| **sharing** | `{ lockedBy?: string }` | optional | View sharing configuration: who locked the view configuration |
| **requiredPermissions** | `string[]` | optional | [ADR-0066] Capabilities a user must ALL hold for this list view or dashboard to be served to them — names that permission sets grant through `systemPermissions`, the same list as `requiredPermissions` on an app, a navigation item or an action. Absent or empty: no audience gate. There is no any-of form: to serve one item to several audiences (legal OR admin), declare one capability and grant it to each of their permission sets. The server enforces it: a user who does not hold them all is not listed the item and is refused it by name; a view container's list views are pruned the same way, and so are an object's own list views for a user who may not edit the object. |
| **rowHeight** | `Enum<'compact' \| 'short' \| 'medium' \| 'tall' \| 'extra_tall'>` | optional | Row height / density setting |
| **grouping** | `{ fields: object[] }` | optional | Group records by one or more fields — server-side: the groups and their header numbers come from an aggregate query over the whole filtered set, rows within a group are paged (see GroupingConfigSchema) |
| **rowColor** | `{ field: string; colors?: Record<string, string> }` | optional | Color rows based on field value |
| **hiddenFields** | `string[]` | optional | Fields to hide in this specific view — the SUBTRACTION of the per-view field composition `columns` x `hiddenFields` x `fieldOrder`: each name listed here is removed from the `columns` projection before `fieldOrder` orders the remainder. A name `columns` never projected subtracts nothing. |
| **fieldOrder** | `string[]` | optional | Explicit field display order for this view — the ORDERING of the per-view field composition `columns` x `hiddenFields` x `fieldOrder`: it sorts what survives `columns` minus `hiddenFields` and never adds a field. A surviving column absent from `fieldOrder` sorts LAST, after every listed one, keeping its `columns`-relative order; a name listed here that did not survive orders nothing. |
| **rowActions** | `string[]` | optional | Actions available for individual row items |
| **bulkActions** | `string[]` | optional | Actions available when multiple rows are selected |
| **bulkActionDefs** | `{ name: string; label?: string; icon?: string; variant?: Enum<'primary' \| 'secondary' \| 'danger' \| 'ghost' \| 'outline'>; … }[]` | optional | Rich bulk action definitions (schema-driven, executed via BulkActionDialog). Use a def for a mass data-plane mutation ('update' with a `patch` / 'delete') that no action expresses, or for an `operation: 'custom'` + `execution: 'aggregate'` entry that dispatches the action it NAMES once for the whole selection — the renderer injects `params._selectedIds: string[]` (read that on the server, not `recordId`) so a single call can produce one aggregate artifact (zip of QR codes, merged PDF, batch print). Aggregate results are all-or-nothing: a handler that cannot cover the whole selection must reject, and per-row retry is replaced by re-running the action. `batchSize` does not apply (the call is never chunked); set `maxRecords` on defs whose server work is expensive. For the PER-RECORD dispatch use `bulkActions: ['<name>']` instead — the bare-string form, promoted with the action's own label, params and `visible`; a 'custom' def without `execution: 'aggregate'` has no dispatcher and is refused at parse time. Toolbar url/api actions can also interpolate the current selection via `${ctx.selection.ids}` / `${ctx.selection.count}`. |
| **conditionalFormatting** | `{ condition: string \| object; style: Record<string, string> }[]` | optional | Conditional formatting rules for list rows |
| **inlineEdit** | `boolean` | optional | Allow inline editing of records directly in the list view |
| **exportOptions** | `Enum<'csv' \| 'xlsx' \| 'json'>[] \| { formats?: Enum<'csv' \| 'xlsx' \| 'json'>[]; maxRecords?: integer; includeHeaders?: boolean; fileNamePrefix?: string; … }` | optional | Export configuration for the list toolbar export menu: `{ formats?, maxRecords?, includeHeaders?, fileNamePrefix?, streaming? }`. A bare format array is the legacy spelling and lifts to `{ formats: [...] }` at parse. |
| **userActions** | `{ sort?: boolean; search?: boolean; filter?: boolean; refresh?: boolean; … }` | optional | User action toggles for the view toolbar |
| **appearance** | `{ showDescription?: boolean; allowedVisualizations?: Enum<'grid' \| 'kanban' \| 'gallery' \| 'calendar' \| 'timeline' \| 'gantt' \| 'map' \| 'chart' \| 'tree'>[] }` | optional | Appearance and visualization configuration |
| **tabs** | `never` | optional | [REMOVED] `view.list.tabs` was removed in @objectstack/spec 17.5.0 (ADR-0049 enforce-or-remove) — no renderer ever mounted a tab bar for it, so authoring it drew nothing: the tab strip above an object's records is the saved-view switcher (ViewTabBar), which renders one tab per named list view and never read this key. Delete the key, and move each tab you want to a named list view under the object's `listViews` instead: the tab's `name` becomes the entry's key, its `label` the entry's `label`, and its `filter` rules join the view's own `filter` on that entry (copy the view's `columns` too); a tab whose `view` already named a list view needs nothing more. Every `listViews` entry renders as a tab in the switcher. Run `os migrate meta --from 17` to list the mechanical edits for existing sources; `--write` applies the ones it can prove, and you apply the rest by hand. |
| **addRecord** | `{ enabled: boolean; position?: Enum<'top' \| 'bottom' \| 'both'> }` | optional | Add record entry point configuration |
| **showRecordCount** | `boolean` | optional | Show record count at the bottom of the list |
| **allowPrinting** | `boolean` | optional | Allow users to print the view |
| **emptyState** | `{ title?: string \| Record<string, string>; message?: string \| Record<string, string>; icon?: string; link?: object }` | optional | Empty state configuration when no records found |
| **aria** | `{ ariaLabel?: string \| Record<string, string>; ariaDescribedBy?: string; role?: string }` | optional | ARIA accessibility attributes for the list view |
| **userFilters** | `{ element?: Enum<'dropdown' \| 'toggle'>; fields?: object[] }` | optional |  |

### Nested Shape: `GetUiViewResponse.formViews[string]`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **type** | `Enum<'simple' \| 'tabbed' \| 'wizard'>` | optional (default: `"simple"`) | Form layout: 'simple' (stacked sections, the default), 'tabbed' (one tab per section) or 'wizard' (one gated step per section). Where the form opens is the caller's choice, not the form's |
| **layout** | `Enum<'vertical' \| 'horizontal'>` | optional | Field layout direction — 'vertical' (the renderer default) or 'horizontal'. Multi-column is not a layout value: set `columns` |
| **columns** | `integer` | optional | Number of columns for the form body |
| **title** | `string` | optional | Form title |
| **description** | `string` | optional | Form description |
| **defaultTab** | `string` | optional | Initially active tab (tabbed forms) |
| **tabPosition** | `Enum<'top' \| 'bottom' \| 'left' \| 'right'>` | optional | Tab strip position (tabbed forms) |
| **allowSkip** | `boolean` | optional | Wizard step-gate opt-out: allow entering a later step without submitting the one before it. Navigation freedom, NOT a validation exemption — the final submit still re-validates every step and returns to the first failing one. Default (absent/false): steps unlock in array order as each prior step submits validly. UI admission only, never authorization. |
| **showStepIndicator** | `boolean` | optional | Show the wizard step indicator (renderer default: shown). Step progress (completed/current/upcoming/invalid) is derived from the step gate — this boolean is the only authorable progress surface. |
| **splitDirection** | `never` | optional | [REMOVED] `view.form.splitDirection` was removed in @objectstack/spec 18.3.0 (ADR-0050) — it set the pane orientation of a split form, and the split form renderer retired in the same release, so no form draws panes. Delete the key. Run `os migrate meta --from 17` to list the mechanical edits for existing sources; `--write` applies the ones it can prove, and you apply the rest by hand. |
| **splitSize** | `never` | optional | [REMOVED] `view.form.splitSize` was removed in @objectstack/spec 18.3.0 (ADR-0050) — it sized the primary pane of a `type: 'split'` form, and `split` left the form view `type` in the same release, so no form view draws panes. Delete the key. Run `os migrate meta --from 17` to list the mechanical edits for existing sources; `--write` applies the ones it can prove, and you apply the rest by hand. |
| **splitResizable** | `never` | optional | [REMOVED] `view.form.splitResizable` was removed in @objectstack/spec 18.3.0 (ADR-0050) — it made the panes of a `type: 'split'` form resizable, and `split` left the form view `type` in the same release, so no form view draws panes. Delete the key. Run `os migrate meta --from 17` to list the mechanical edits for existing sources; `--write` applies the ones it can prove, and you apply the rest by hand. |
| **drawerSide** | `never` | optional | [REMOVED] `view.form.drawerSide` was removed in @objectstack/spec 18.3.0 (ADR-0049) — no form view opens in a drawer, so no form view ever read it. Delete the key: a list opens a clicked record in its drawer (`navigation.mode`), and the React `ObjectForm` block keeps its own `drawerSide` prop for the drawer it opens with `formType='drawer'`. Run `os migrate meta --from 17` to list the mechanical edits for existing sources; `--write` applies the ones it can prove, and you apply the rest by hand. |
| **drawerWidth** | `never` | optional | [REMOVED] `view.form.drawerWidth` was removed in @objectstack/spec 18.3.0 (ADR-0049) — no form view opens in a drawer, so no form view ever read it, and a pixel width cannot be chosen without knowing the client viewport. Delete the key: a list's record drawer takes its starting width from the `navigation.size` bucket, and the React `ObjectForm` block keeps its own `drawerWidth` prop for the drawer it opens. Run `os migrate meta --from 17` to list the mechanical edits for existing sources; `--write` applies the ones it can prove, and you apply the rest by hand. |
| **modalSize** | `Enum<'sm' \| 'default' \| 'lg' \| 'xl' \| 'full'>` | optional | Size bucket of the dialog a caller opens this form in: the console's New / Edit dialog, or the modal container of the React `ObjectForm` block (formType 'modal') |
| **data** | `{ provider: 'object'; object: string } \| { provider: 'value'; items: any[] } \| { provider: 'schema'; schemaId: string; schema?: Record<string, any> }` | optional | Data source configuration (defaults to "object" provider) |
| **sections** | `{ name?: string; label?: string \| Record<string, string>; description?: string; collapsible?: boolean; … }[]` | optional |  |
| **groups** | `{ name?: string; label?: string \| Record<string, string>; description?: string; collapsible?: boolean; … }[]` | optional | [LEGACY ALIAS → `sections`] Accepted for back-compat and folded onto `sections` at parse; `sections` wins when both are present. Prefer `sections`. |
| **subforms** | `{ childObject: string; relationshipField?: string; columns?: object[]; amountField?: string; … }[]` | optional | Inline master-detail child collections |
| **sharing** | `{ enabled?: boolean; publicLink?: string; password?: string; allowedDomains?: string[]; … }` | optional | Public sharing configuration for this form |
| **submitBehavior** | `{ kind: 'thank-you'; title?: string; message?: string } \| { kind: 'redirect'; url: string; delayMs?: integer } \| { kind: 'continue' } \| { kind: 'next-record' }` | optional | Post-submit behavior. On the `redirect` arm, `url` is relative-only and interpolates only declared record fields as `{{record.field_name}}`, URL-escaped. |
| **buttons** | `{ submit?: object; cancel?: object }` | optional | Form action-button visibility & labels; folded onto the flat renderer props by ObjectUI ObjectForm. |
| **defaults** | `Record<string, any>` | optional | Initial field values for create-mode forms (folded into ObjectUI ObjectForm initial values;). |

### Nested Shape: `GetUiViewResponse.protection`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **lock** | `Enum<'none' \| 'no-overlay' \| 'no-delete' \| 'full'>` | ✅ | Lock policy — none \| no-overlay \| no-delete \| full. |
| **reason** | `string` | ✅ | User-visible reason shown when the lock blocks an action. |
| **docsUrl** | `string` | optional | Optional URL the Studio banner links to for more context. |


---

## HistoryMetaItemRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **type** | `string` | ✅ | Metadata type name |
| **name** | `string` | ✅ | Item name |
| **sinceSeq** | `number` | optional | Exclusive lower bound on `seq` for pagination: only events with `seq > sinceSeq` are returned. Absent means "from the beginning". |
| **limit** | `number` | optional | Maximum events to return, oldest first. Forwarded to the repository unclamped and with NO default — absent means the full remaining change log. (Deliberately no declared bounds: unlike the audit twin's [1, 500] clamp, nothing on this path clamps or refuses, so declaring `.min()`/`.max()` here would refuse values the shipped verb accepts.) |
| **member** | `string` | optional | Bundle member selector, for a metadata type whose identity is a pair (today only `email_template`, identified by `(name, locale)`): the member this request addresses, spelled as its discriminator value (`zh-CN`). Absent, the canonical member (`en-US` for `email_template`), which is what a bare-name read answers; naming the canonical value is the same request. It is not the response translation locale, which `?locale=` and `Accept-Language` choose. Refused with 400 `VALIDATION_ERROR` on a type that has no members, and for a blank value or one longer than a stored member can be (100 characters). A well-formed member with nothing stored or shipped answers as that door answers an absent item. |


---

## HistoryMetaItemResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **events** | `{ seq: integer; op: Enum<'create' \| 'update' \| 'delete' \| 'rename' \| 'publish' \| 'revert'>; ref: object; hash: string \| null; … }[]` | ✅ | The durable change-log for the item, oldest first. See the schema-level note for what an empty array means — and what it never means. |

### Nested Shape: `HistoryMetaItemResponse.events[number]`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **seq** | `integer` | ✅ | Sequence number this write produced in the org log (sys_metadata_history.event_seq) — the token the request's `sinceSeq` filters on, and the same key the write-verb receipts (`SaveMetaItemResponseSchema.seq` and siblings) carry. |
| **op** | `Enum<'create' \| 'update' \| 'delete' \| 'rename' \| 'publish' \| 'revert'>` | ✅ | Which change-log operation the event records (ADR-0008 §2.4). Closed vocabulary, mirrored from the producer's own enum. |
| **ref** | `{ org: string; type: string; name: string; version?: string }` | ✅ | Which item the event is about. |
| **hash** | `string \| null` | ✅ | Content hash of the body this event wrote; `null` when the event wrote none (`op="delete"`). |
| **parentHash** | `string \| null` | ✅ | Hash the written version was derived from; `null` for a first version. |
| **version** | `integer` | optional | Per-(org,type,name) monotonic lineage counter at this event — the token `rollbackMetaItem({ toVersion })` pins. Absent when the row recorded none. |
| **previousName** | `string` | optional | Set on op="rename": the old machine name. |
| **actor** | `string \| null` | ✅ | Who wrote this. `null` = system-initiated (boot sync, migration, scheduled job) — never a sentinel string: consumers that resolve this against `sys_user` must be able to tell "nobody" from "a user id". |
| **message** | `string` | optional | Optional commit message recorded with the write. |
| **ts** | `string` | ✅ | When the write happened (ISO-8601 string). |
| **source** | `string` | ✅ | Which code path wrote the event: an open label the server stamps, not an enum an author or client picks from (e.g. `protocol.saveMetaItem`, `protocol.publishMetaItem`, `migrate-stored`). The history row's stored value is passed through; `sys-metadata-repo` is the repository's own default and the answer for a row that stored none. |


---

## HttpFindQueryParams

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **filter** | `string` | optional | JSON-encoded filter expression (canonical, singular). |
| **filters** | `string` | optional | JSON-encoded filter expression (deprecated plural alias). |
| **select** | `string` | optional | Comma-separated list of fields to retrieve — at least one field name; a value naming no field (blank, or only commas and spaces) is refused. |
| **sort** | `string` | optional | Sort expression (e.g. "name asc,created_at desc" or "-created_at"). |
| **orderBy** | `string` | optional | Alias for sort (OData compatibility). |
| **top** | `integer \| string` | optional | Max records to return (limit) — a non-negative integer. |
| **skip** | `integer \| string` | optional | Records to skip (offset) — a non-negative integer. |
| **expand** | `string` | optional | Comma-separated list of lookup/master_detail field names to expand. Resolved to populate array and passed to the engine for batch $in expansion. |
| **search** | `string` | optional | Full-text search query. |
| **distinct** | `never` | optional | [REMOVED] `query.distinct` was removed in @objectstack/spec 17 (ADR-0049 / ADR-0078) — no driver ever rendered SELECT DISTINCT; the flag's only observable effect was MIS-WIRED: the REST list path treated a distinct query as not countable and silently degraded `total`/`hasMore` to a page-local estimate while still returning duplicate rows. Delete the key; `QueryBuilder.distinct()` was removed with it, and the count suppression is gone (`total` is truthful again). For unique values of one column use the SQL/memory drivers' `distinct(object, field)` door; for unique combinations, `groupBy`; for a deduplicated count, the `count_distinct` aggregation. |
| **count** | `boolean \| Enum<'true' \| 'false'>` | optional | Include the total count in the response — `true` or `false`; any other value is refused. |


---

## InstallPackageRequest

Install package request

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **manifest** | `{ id: string; namespace?: string; defaultDatasource?: string; version: string; … }` | ✅ | Package manifest to install |
| **settings** | `Record<string, any>` | optional | User-provided settings at install time |
| **enableOnInstall** | `boolean` | optional | Whether to enable immediately after install — restates the install-door request key, whose one authority is api/PackageInstallRequest; this protocol primitive honours it on the registry row: `true` enables, `false` disables, and ABSENT keeps the row's current lifecycle state (a fresh install lands enabled) |
| **platformVersion** | `string` | optional | Current platform version for compatibility verification |

### Nested Shape: `InstallPackageRequest.manifest`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **id** | `string` | ✅ | Unique package identifier — must match reverse-domain notation (e.g. com.acme.crm) |
| **namespace** | `string` | optional | Short namespace identifier; also the mandatory prefix of every object name (e.g. "todo" → object names "todo_task", "todo_project") |
| **defaultDatasource** | `string` | optional (default: `"default"`) | Default datasource for all objects in this package |
| **version** | `string` | ✅ | Package version (SemVer 2.0.0 — e.g. 1.2.3, 2.0.0-beta.1, 1.0.0+20230101) |
| **type** | `Enum<'plugin' \| 'ui' \| 'driver' \| 'server' \| 'app' \| 'theme' \| 'agent' \| 'objectql' \| …>` | ✅ | Type of package |
| **scope** | `Enum<'cloud' \| 'system' \| 'project'>` | optional (default: `"project"`) | Deployment scope: cloud \| system \| project |
| **installModes** | `Enum<'managed' \| 'unmanaged'>[]` | optional | Install modes this package permits (ADR-0131 D6): managed \| unmanaged. Absent means ['managed']. |
| **name** | `string` | ✅ | Human-readable package name |
| **description** | `string` | optional | Package description |
| **permissions** | `{ services?: string[]; hooks?: string[]; network?: string[]; fs?: string[] }` | optional | Required permissions at the AUTHORING stage: the structured plugin block `{ services, hooks, network, fs }` (ADR-0025 §3.2) — the legacy flat string list is retired — and at the assembled stage the same key is the ADR-0090 `PermissionSet[]` collection instead (`AssembledPackageBodySchema`) |
| **objects** | `string[]` | optional | Glob patterns for ObjectQL schemas files |
| **datasources** | `string[]` | optional | Glob patterns for Datasource definitions |
| **dependencies** | `Record<string, string>` | optional | Package dependencies |
| **configuration** | `never` | optional | [REMOVED] `manifest.configuration` was removed in @objectstack/spec 17 (ADR-0049 enforce-or-remove) — nothing ever read the block: no settings UI rendered it and no loader resolved a setting from it, so authoring it configured nothing. Worse, `properties.*.secret` promised "value is encrypted/masked (e.g. API Keys)" while nothing encrypted, masked or even parsed the flag — a false assurance about credential handling. Delete the key. A plugin is configured by the host that composes it: pass options to its constructor in `defineStack({ plugins: [new MyPlugin({ … })] })`, which is the enforced channel. A declarative settings surface must be designed with an enforcing reader first, not revived here. |
| **contributes** | `{ kinds?: object[] }` | optional | Platform contributions |
| **data** | `{ object: string; externalId?: string \| string[]; mode?: Enum<'insert' \| 'update' \| 'upsert' \| 'replace' \| 'ignore'>; columns?: Record<string, object>; … }[]` | optional | Initial seed data (prefer top-level data field) |
| **capabilities** | `never` | optional | [REMOVED] `manifest.capabilities` was removed in @objectstack/spec 17 (ADR-0049 enforce-or-remove) — no discovery path ever consulted the block: nothing read `implements`, `provides`, `requires`, `extensionPoints` or `extensions`, so the declared "interoperability and automatic discovery" never happened. Delete the key. Real dependency resolution runs off top-level `manifest.dependencies`, which stays. Capability-based discovery must be designed with an enforcing reader first, not revived here. |
| **extensions** | `never` | optional | [REMOVED] `manifest.extensions` was removed in @objectstack/spec 17 (ADR-0049 enforce-or-remove) — an untyped map with zero readers: whatever was parked here was stored and never consulted. Delete the key. Extend the platform through the enforced channels instead: `contributes.kinds` registers metadata kinds, `navigationContributions` injects navigation into other packages' apps, and code-level extension happens in the plugin itself (`init`/`start`). |
| **navigationContributions** | `{ app: string; group?: string; priority?: integer; items: (object \| … +9 more)[] }[]` | optional | Navigation items this package contributes into apps owned by other packages |
| **engine** | `{ objectstack: string }` | optional | Platform compatibility requirements (legacy; superseded by `engines`) |
| **engines** | `{ platform?: string; protocol?: string }` | optional | Plugin compatibility ranges (ADR-0025 §3.2; supersedes `engine`) |
| **runtime** | `Enum<'node' \| 'sandbox' \| 'worker'>` | optional | Plugin trust tier the plugin declares (ADR-0025 §3.6) — enforced at the cloud marketplace publish gate (unverified publisher requesting `node` → HTTP 422 + manual review) and at boot for a plugin `os plugin install` installed: only `node` is loaded (in process, not isolated); `sandbox` and `worker` are refused until ADR-0025 M3 routes them |
| **packaging** | `Enum<'bundled'>` | optional | Dependency packaging strategy (ADR-0025 v2 §3.3): `bundled`, the only one and the default — omit the key or write `bundled` |
| **main** | `string` | optional | Entry module of a code-bearing plugin, relative to the plugin root; `os plugin build` bundles it and writes `dist/index.mjs` here in the compiled manifest (ADR-0025 §3.4) |
| **integrity** | `Record<string, string>` | optional | Per-file content digests of the plugin artifact (ADR-0025 §3.2) |


---

## InstallPackageResponse

Install package response

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **package** | `{ manifest: object; source: Enum<'code' \| 'marketplace' \| 'authored'>; status?: Enum<'installed' \| 'disabled' \| 'installing' \| 'upgrading' \| 'uninstalling' \| 'error'>; enabled?: boolean; … }` | ✅ | Installed package details |
| **message** | `string` | optional | Installation status message |
| **dependencyResolution** | `{ dependencies: object[]; canProceed: boolean; requiredActions: object[]; installOrder: string[]; … }` | optional | Dependency resolution result from install analysis |

### Nested Shape: `InstallPackageResponse.package`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **manifest** | `{ id: string; namespace?: string; defaultDatasource?: string; version: string; … }` | ✅ | Package manifest at the AUTHORING stage; a row installed by a `defineStack()` host carries the assembled body instead — see `AssembledInstalledPackageSchema` / `InstalledPackageAtEitherStageSchema` |
| **source** | `Enum<'code' \| 'marketplace' \| 'authored'>` | ✅ | Where the installed package came from: code (registered by the deployment's own code), marketplace (installed from the marketplace), or authored (a database-authored package) |
| **status** | `Enum<'installed' \| 'disabled' \| 'installing' \| 'upgrading' \| 'uninstalling' \| 'error'>` | optional (default: `"installed"`) | Package state: installed, disabled, installing, upgrading, uninstalling, or error |
| **enabled** | `boolean` | optional (default: `true`) | Whether the package is currently enabled |
| **installedAt** | `string` | optional | Installation timestamp |
| **updatedAt** | `string` | optional | Last update timestamp |
| **installedVersion** | `string` | optional | Currently installed version for quick access |
| **previousVersion** | `string` | optional | Version before the last upgrade |
| **statusChangedAt** | `string` | optional | Status change timestamp |
| **errorMessage** | `string` | optional | Error message when status is error |
| **settings** | `Record<string, any>` | optional | User-provided configuration settings |
| **upgradeHistory** | `{ fromVersion: string; toVersion: string; upgradedAt: string; status: Enum<'success' \| 'failed' \| 'rolled_back'>; … }[]` | optional | Version upgrade history |
| **registeredNamespaces** | `string[]` | optional | Namespace prefixes registered by this package |

### Nested Shape: `InstallPackageResponse.dependencyResolution`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **dependencies** | `{ packageId: string; requiredRange: string; resolvedVersion?: string; installedVersion?: string; … }[]` | ✅ | Resolution result for each dependency |
| **canProceed** | `boolean` | ✅ | Whether installation can proceed |
| **requiredActions** | `{ type: Enum<'install' \| 'upgrade' \| 'confirm_conflict'>; packageId: string; description: string }[]` | ✅ | Actions required before proceeding |
| **installOrder** | `string[]` | ✅ | Topologically sorted package IDs for installation |
| **circularDependencies** | `string[][]` | optional | Circular dependency chains detected (e.g. [["A", "B", "A"]]) |


---

## ListAiConversationsRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **agentId** | `string` | optional | Filter by agent |
| **limit** | `integer` | optional | Maximum conversations to return |
| **cursor** | `string` | optional | The `id` of the last conversation on the previous page. The next page starts with the conversation created immediately before it, continuing newest first. Omit it to read the first page. An id that names no conversation of the caller is refused rather than read as the start of the list. |


---

## ListAiConversationsResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **conversations** | `{ id: string; title?: string; agentId?: string; userId?: string; … }[]` | ✅ | The caller's conversations, newest first — ordered by creation time, then `id`, both descending |
| **hasMore** | `boolean` | ✅ | Whether at least one more conversation follows this page. When `true`, send the `id` of the last conversation in `conversations` as `cursor` to read the next page. |

### Nested Shape: `ListAiConversationsResponse.conversations[number]`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **id** | `string` | ✅ | Conversation id |
| **title** | `string` | optional | Title / summary |
| **agentId** | `string` | optional | Agent this conversation is bound to |
| **userId** | `string` | optional | Owning user |
| **messages** | `({ role: Enum<'system' \| 'user' \| 'assistant' \| 'tool'>; content?: any; parts?: any[] } & Record<string, any>)[]` | ✅ | Message history |
| **createdAt** | `string` | ✅ | Creation timestamp (ISO 8601) |
| **updatedAt** | `string` | ✅ | Last update timestamp (ISO 8601) |
| **metadata** | `Record<string, any>` | optional | Conversation metadata |


---

## ListAiPendingActionsRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **status** | `Enum<'pending' \| 'approved' \| 'executed' \| 'failed' \| 'rejected'>` | optional | Filter by status |
| **conversationId** | `string` | optional | Filter by proposing conversation |
| **limit** | `integer` | optional | Max rows (server default 100) |


---

## ListAiPendingActionsResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **items** | `{ id: string; object_name: string; action_name: string; tool_name: string; … }[]` | ✅ | Queued actions, newest first |
| **total** | `number` | ✅ | Number of rows returned (not a total across pages) |

### Nested Shape: `ListAiPendingActionsResponse.items[number]`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **id** | `string` | ✅ | Pending action id |
| **object_name** | `string` | ✅ | Object the action targets |
| **action_name** | `string` | ✅ | Action name |
| **tool_name** | `string` | ✅ | Tool that would execute it |
| **tool_input** | `string` | ✅ | Serialized tool input |
| **status** | `Enum<'pending' \| 'approved' \| 'executed' \| 'failed' \| 'rejected'>` | ✅ | Lifecycle status |
| **result** | `string` | optional | Serialized result, once executed |
| **error** | `string` | optional | Failure message, when status is failed |
| **rejection_reason** | `string` | optional | Reason given at rejection |
| **conversation_id** | `string` | optional | Conversation that proposed it |
| **message_id** | `string` | optional | Message that proposed it |
| **proposed_by** | `string` | optional | Actor that proposed it |
| **decided_by** | `string` | optional | Actor that approved or rejected it |
| **proposed_at** | `string` | ✅ | Proposal timestamp (ISO 8601) |
| **decided_at** | `string` | optional | Decision timestamp (ISO 8601) |


---

## ListDraftsResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **drafts** | `{ type: string; name: string; member?: string; label: string \| Record<string, string> \| null; … }[]` | ✅ | Every pending draft visible to the caller, one row per item. |

### Nested Shape: `ListDraftsResponse.drafts[number]`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **type** | `string` | ✅ | Metadata type name (canonical singular). |
| **name** | `string` | ✅ | Item name. |
| **member** | `string` | optional | The bundle member this draft is, for a metadata type whose identity is a pair (today only `email_template`, identified by `(name, locale)`): its discriminator value (`zh-CN`), and the canonical member spelled out (`en-US`). Present on every draft of such a type, so two members of one name are two entries; absent for a type whose identity is its name alone. Pass it back as the by-name doors' `member` selector (`?member=`) to read, publish or discard this draft. |
| **label** | `string \| Record<string, string> \| null` | ✅ | The draft body's own top-level `label`, as authored: a plain string or an inline locale map (`I18nLabel`), resolved by the reader. `null` when the body declares none, or none this shape admits — never the item name standing in for it. |
| **packageId** | `string \| null` | ✅ | Package the draft is bound to, `null` for a package-less draft. |
| **updatedAt** | `string \| null` | ✅ | Last-touch timestamp of the draft row (ISO-8601 string), `null` when the row recorded none. |
| **updatedBy** | `string \| null` | ✅ | Who last touched the draft, `null` when the row recorded none. |


---

## ListNotificationsRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **read** | `boolean` | optional | Filter by read status |
| **type** | `string` | optional | Filter by notification type |
| **limit** | `number` | optional | Maximum number of notifications to return — the newest N. Omitted leaves the window to the server, which is not a fixed part of this contract: the platform inbox answers 50 and clamps any requested value into 1..200 rather than refusing it. This endpoint is not paginated — there is no continuation token, so a larger window is the only way to see more. |
| **cursor** | `never` | optional | [REMOVED] `cursor` was removed from GET /api/v1/notifications in @objectstack/spec 17 (ADR-0049) — it was declared on the request AND the response and honoured on neither: the server reads only `read`/`type`/`limit`, and no emit site ever wrote the response key, so a caller paginating by it re-read the first window forever with no error and no 400. Delete the key; the `cursor` argument of `client.notifications.list()` was removed with it. This route is NOT paginated — it answers the newest `limit` notifications and stops, so ask for a bigger window (`limit`, clamped by the server into 1..200) instead of a next page. A first-class inbox cursor, if ever built, will be a response-minted opaque token, not this key. |


---

## ListNotificationsResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **notifications** | `{ id: string; type: string; title: string; body: string; … }[]` | ✅ | List of notifications — the newest window, not a page |
| **unreadCount** | `number` | ✅ | Total number of unread notifications |
| **cursor** | `never` | optional | [REMOVED] `cursor` was removed from GET /api/v1/notifications in @objectstack/spec 17 (ADR-0049) — it was declared on the request AND the response and honoured on neither: the server reads only `read`/`type`/`limit`, and no emit site ever wrote the response key, so a caller paginating by it re-read the first window forever with no error and no 400. Delete the key; the `cursor` argument of `client.notifications.list()` was removed with it. This route is NOT paginated — it answers the newest `limit` notifications and stops, so ask for a bigger window (`limit`, clamped by the server into 1..200) instead of a next page. A first-class inbox cursor, if ever built, will be a response-minted opaque token, not this key. |

### Nested Shape: `ListNotificationsResponse.notifications[number]`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **id** | `string` | ✅ | Notification ID |
| **type** | `string` | ✅ | Notification type |
| **title** | `string` | ✅ | Notification title |
| **body** | `string` | ✅ | Notification body text |
| **read** | `boolean` | optional (default: `false`) | Whether notification has been read |
| **data** | `Record<string, any>` | optional | Additional notification data |
| **actionUrl** | `string` | optional | URL to navigate to when clicked |
| **createdAt** | `string` | ✅ | When notification was created |


---

## ListPackagesRequest

List packages request

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **status** | `Enum<'installed' \| 'disabled' \| 'installing' \| 'upgrading' \| 'uninstalling' \| 'error'>` | optional | Filter by package status |
| **type** | `Enum<'plugin' \| 'ui' \| 'driver' \| 'server' \| 'app' \| 'theme' \| 'agent' \| 'objectql' \| 'module' \| 'gateway' \| 'adapter'>` | optional | Filter by package type |
| **enabled** | `boolean` | optional | Filter by enabled state |


---

## ListPackagesResponse

List packages response

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **packages** | `{ manifest: object; source: Enum<'code' \| 'marketplace' \| 'authored'>; status?: Enum<'installed' \| 'disabled' \| 'installing' \| 'upgrading' \| 'uninstalling' \| 'error'>; enabled?: boolean; … }[]` | ✅ | List of installed packages |
| **total** | `number` | ✅ | Total package count |

### Nested Shape: `ListPackagesResponse.packages[number]`

Installed package with runtime lifecycle state

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **manifest** | `{ id: string; namespace?: string; defaultDatasource?: string; version: string; … }` | ✅ | Package manifest at the AUTHORING stage; a row installed by a `defineStack()` host carries the assembled body instead — see `AssembledInstalledPackageSchema` / `InstalledPackageAtEitherStageSchema` |
| **source** | `Enum<'code' \| 'marketplace' \| 'authored'>` | ✅ | Where the installed package came from: code (registered by the deployment's own code), marketplace (installed from the marketplace), or authored (a database-authored package) |
| **status** | `Enum<'installed' \| 'disabled' \| 'installing' \| 'upgrading' \| 'uninstalling' \| 'error'>` | optional (default: `"installed"`) | Package state: installed, disabled, installing, upgrading, uninstalling, or error |
| **enabled** | `boolean` | optional (default: `true`) | Whether the package is currently enabled |
| **installedAt** | `string` | optional | Installation timestamp |
| **updatedAt** | `string` | optional | Last update timestamp |
| **installedVersion** | `string` | optional | Currently installed version for quick access |
| **previousVersion** | `string` | optional | Version before the last upgrade |
| **statusChangedAt** | `string` | optional | Status change timestamp |
| **errorMessage** | `string` | optional | Error message when status is error |
| **settings** | `Record<string, any>` | optional | User-provided configuration settings |
| **upgradeHistory** | `{ fromVersion: string; toVersion: string; upgradedAt: string; status: Enum<'success' \| 'failed' \| 'rolled_back'>; … }[]` | optional | Version upgrade history |
| **registeredNamespaces** | `string[]` | optional | Namespace prefixes registered by this package |


---

## MarkAllNotificationsReadRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |


---

## MarkAllNotificationsReadResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **success** | `boolean` | ✅ | Whether the operation succeeded |
| **readCount** | `number` | ✅ | Number of notifications marked as read |


---

## MarkNotificationsReadRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **ids** | `string[]` | ✅ | Notification IDs to mark as read |


---

## MarkNotificationsReadResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **success** | `boolean` | ✅ | Whether the operation succeeded |
| **readCount** | `number` | ✅ | Number of notifications marked as read |


---

## MetadataAuthoringClient

The OAuth client an MCP build-door write was made by: the client a person consented to with `metadata:write`, as the MCP build door verified it from the request's token. Closed: a member other than `id` and `name` is refused, never stripped.

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **id** | `string` | ✅ | The OAuth client id the verified token was issued to. Opaque — compare it, never parse it. |
| **name** | `string` | optional | The client's registered display name (its OAuth `client_name`) when it registered one, as registered. Absent when it registered none; ⛔ never the id standing in for it. |


---

## MetadataAuthoringSurface

The authoring surface of a metadata write. Closed: `copilot` is an AI assistant's tool writing on a person's behalf (Studio's AI copilot); `mcp` is an AI agent outside the console writing on a person's behalf through the MCP build door, as the OAuth client the person consented to; `studio` is a person editing in Studio; `api` is any other caller of the save door. It names the surface, not the person: a copilot write, an MCP write and a Studio write of the same user carry the same `actor`.

### Allowed Values

* `copilot`
* `mcp`
* `studio`
* `api`


---

## MetadataConflictError

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **error** | `string` | ✅ | Human-readable refusal sentence, unchanged by `currentVersion`: it names the item and the version the door judged current. For display and logs; branch on `code` and read `currentVersion`, never parse this. |
| **code** | `'METADATA_CONFLICT'` | ✅ | ADR-0112 registered code of the optimistic-lock refusal. |
| **currentVersion** | `string \| null` | ✅ | The version token of the head the door judged current — the value the sentence names — or `null` when no row of the target lifecycle exists. Send it as `If-Match` to overwrite deliberately (or `If-None-Match: *` when it is `null`), or re-read the item to merge first. A `null` for an item whose read serves `editable: false` prescribes no pin: that item takes no save. Opaque — echo it verbatim, never parse it. |


---

## Notification

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **id** | `string` | ✅ | Notification ID |
| **type** | `string` | ✅ | Notification type |
| **title** | `string` | ✅ | Notification title |
| **body** | `string` | ✅ | Notification body text |
| **read** | `boolean` | optional (default: `false`) | Whether notification has been read |
| **data** | `Record<string, any>` | optional | Additional notification data |
| **actionUrl** | `string` | optional | URL to navigate to when clicked |
| **createdAt** | `string` | ✅ | When notification was created |


---

## NotificationPreferences

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **email** | `boolean` | optional (default: `true`) | Receive email notifications |
| **push** | `boolean` | optional (default: `true`) | Receive push notifications |
| **inApp** | `boolean` | optional (default: `true`) | Receive in-app notifications |
| **digest** | `Enum<'none' \| 'daily' \| 'weekly'>` | optional (default: `"none"`) | Email digest frequency |
| **channels** | `Record<string, { enabled?: boolean; email?: boolean; push?: boolean }>` | optional | Per-channel notification preferences |

### Nested Shape: `NotificationPreferences.channels[string]`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **enabled** | `boolean` | optional (default: `true`) | Whether this channel is enabled |
| **email** | `boolean` | optional | Override email setting |
| **push** | `boolean` | optional | Override push setting |


---

## PublicFormSubmitResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **id** | `string` | ✅ | The ID of the newly created record. |
| **fields** | `Record<string, string \| number>` | optional | The stored values of exactly the fields the form's `thank-you` panel names as `{{record.field_name}}`, keyed by field name — each one a field the engine generates on insert (an `autonumber` field, e.g. a case number). Present only when the panel names at least one; never any other field of the record. |


---

## PublishMetaItemRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **type** | `string` | ✅ | Metadata type name |
| **name** | `string` | ✅ | Item name — lowercase snake_case segments, optionally dot-qualified (`crm_lead`, `crm_lead.pipeline`). The promotion door enforces the same grammar as `saveMetaItem`. |
| **actor** | `string` | optional | Identity recorded on the `op='publish'` history event. On the REST door this is the request's authenticated identity (one producer) — never a caller-supplied header. |
| **message** | `string` | optional | Optional human-readable note recorded with the publish history event. |
| **packageId** | `string \| null` | optional | ADR-0048 — the software package the draft being promoted was listed under, when the caller has one to state (`?package=<id>` on the REST door). Named, the publish promotes that package's draft and no other. ABSENT names no package: the publish promotes the draft a draft save that names no package writes — the draft of the one package the item's active rows are bound to, else its package-unbound draft, promoted into that package — and answers `404 NO_DRAFT` when neither exists, as the strict draft read does. For an item never published into a package it is the package-unbound draft, else the one package's draft. It is refused `409 RESOURCE_CONFLICT` when the active rows (or, with none and no package-unbound draft, the drafts) are bound to two or more packages. ⚠️ `null` is NOT the same as absent: the implementation branches on the KEY BEING PRESENT, and `null` pins the lookup to the package-UNBOUND row. Spread it in conditionally; a present-and-`undefined` key coerces to `null` downstream and makes a package-bound draft unfindable — a silent `no_draft`. |
| **member** | `string` | optional | Bundle member selector, for a metadata type whose identity is a pair (today only `email_template`, identified by `(name, locale)`): the member this request addresses, spelled as its discriminator value (`zh-CN`). Absent, the canonical member (`en-US` for `email_template`), which is what a bare-name read answers; naming the canonical value is the same request. It is not the response translation locale, which `?locale=` and `Accept-Language` choose. Refused with 400 `VALIDATION_ERROR` on a type that has no members, and for a blank value or one longer than a stored member can be (100 characters). A well-formed member with nothing stored or shipped answers as that door answers an absent item. |


---

## PublishMetaItemResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **success** | `boolean` | ✅ | Always true on a 2xx — the draft was promoted. It does NOT cover the best-effort side effects below, each of which reports its own `success`. |
| **version** | `string` | ✅ | Version token of the row this promotion left, and the token the ADR-0008 optimistic-concurrency chain runs on: send it back as the `If-Match` request header on the next write to that item and a concurrent edit is refused 409 `METADATA_CONFLICT` instead of silently overwritten. A keyed digest of the stored content hash, never the hash itself — the same form `SaveMetaItemResponseSchema.version` and the item read's `version` serve. Opaque to callers — echo it verbatim, never parse it. Currently emitted as `hmac-sha256:` plus 64 hex characters, but the format is not part of this contract. |
| **seq** | `integer` | ✅ | Monotonic sequence number of the `op='publish'` metadata event this promotion appended to the item history (sys_metadata_history.event_seq). Orders writes; unlike `version` it is not an OCC token. |
| **seedApplied** | `{ success: boolean; inserted: integer; updated: integer; error?: string; … }` | optional | Outcome of materializing a published `seed` body into data rows. Present ONLY when the published type is `seed` — publishing a seed is what makes its rows live, so the load rides along with the metadata promotion. Best-effort: a seed-load problem is surfaced here, never thrown, so a caller must check `seedApplied.success` instead of assuming the 200 covered the data. Absent on the batch path, which suppresses the per-item apply and loads every seed body in one later pass. |
| **materializeApplied** | `{ success: boolean; inserted: integer; updated: integer; error?: string }` | optional | Outcome of the ADR-0086 P2 publish-time materializer — the step that applies the published body to its data-plane state (e.g. `flow`, whose draft credentials become the live ones). Present ONLY when a materializer is registered for this metadata type, which is why it is optional: its absence means "no materializer ran", never "it failed". Best-effort, same contract as `seedApplied`. |
| **projectionApplied** | `{ success: boolean; error?: string }` | optional | Outcome of the awaited ADR-0094 mutation projector — the post-persist step that materializes this metadata into its derived data-plane read model. The same receipt `{@link SaveMetaItemResponseSchema}` carries, because the projector runs on BOTH write doors: a direct active save and this draft→active promotion. Present ONLY when a projector is registered for this metadata type. Best-effort — a projector failure is reported here and logged, never thrown. |
| **advisories** | `{ rule: string; path: string; where: string; message: string; … }[]` | optional | Non-gating findings from the runtime authoring gate — the same shared author-time rules `os validate` / `os build` / `os lint` run, applied to the DRAFT body this promotion carried to `active` (the same key `SaveMetaItemResponseSchema` carries, because the gate runs on both write doors, D1). The promotion SUCCEEDED; these are what the gate has to say about it anyway. Present ONLY when at least one advisory was raised — an empty array is never emitted, so a clean publish's response bytes are unchanged and absence means "nothing to report", never "the gate did not run". Advisory by construction: every entry has `severity` `warning` or `info`, because an `error` finding refuses the promotion and arrives as the 422 `invalid_metadata` envelope instead of here. A caller that ignores this key behaves exactly as before. This door is the one Studio's designer takes on every edit (draft save, then publish), and a Studio / MCP / AI author has no CLI at all — which is the gap this key exists to close. |
| **message** | `string` | optional | Human-readable receipt, e.g. `Published draft — type=view, name=cases [seq=3]`. The producer sets it on every publish today; it stays optional to match the producer's own signature and its `SaveMetaItemResponse` twin, and because an absent human-readable string strips no data — the failure mode this key exists to prevent. |

### Nested Shape: `PublishMetaItemResponse.seedApplied`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **success** | `boolean` | ✅ | False when the seed rows did not fully land. The publish itself still succeeded — check this rather than assuming data went live. |
| **inserted** | `integer` | ✅ | Rows created by the externalId-keyed upsert. |
| **updated** | `integer` | ✅ | Rows updated by the externalId-keyed upsert. |
| **error** | `string` | optional | Single failure message, present when the seed apply threw before the loader ran (including "no readable seed bodies"). |
| **errors** | `any[]` | optional | Per-record failures reported by the seed loader. Present only when the loader ran and returned a non-empty error list. |
| **landed** | `{ object: string; inserted: object; matched: object }[]` | optional | Per object, the stored rows the seed landed on: `inserted` (created by this publish) and `matched` (pre-existing rows a seed key matched), each capped at `SEED_APPLIED_LANDED_IDS_CAP` (200) ids with `total` and `truncated` stating any cut. One entry per object the loader ran, empty lists included. When the load threw part-way (`success: false` with `error`), it lists only the datasets that FINISHED before the throw, and is EMPTY when none had; the dataset that threw is not listed (its rows are in the counts, its object in `error`). ABSENT means "not reported" — the apply failed before any load ran (see `error`), or the producer reports no ids — never "no rows landed". |

### Nested Shape: `PublishMetaItemResponse.materializeApplied`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **success** | `boolean` | ✅ | False when the materializer threw or reported failure; the publish still succeeded. |
| **inserted** | `integer` | ✅ | Data-plane rows created by the materializer. |
| **updated** | `integer` | ✅ | Data-plane rows updated by the materializer. |
| **error** | `string` | optional | Materializer failure message, present only when `success` is false. |

### Nested Shape: `PublishMetaItemResponse.projectionApplied`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **success** | `boolean` | ✅ | False when the projector threw; the metadata promotion itself still succeeded. |
| **error** | `string` | optional | Projector failure message, present only when `success` is false. |

### Nested Shape: `PublishMetaItemResponse.advisories[number]`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **rule** | `string` | ✅ | Stable diagnostic rule id (`flow-multi-write-unfiltered`, `approval-expression-invalid`, …). Machine-readable and stable across releases — the key a renderer groups or suppresses by. |
| **path** | `string` | ✅ | Config path inside the SUBMITTED body (`flows[0].nodes[1].config.multi`), so an editor can jump to the offending key. May be empty when the finding is about the item as a whole. The TOP-LEVEL collection entry follows one rule. The gate evaluates against a private per-write snapshot whose indexes no caller can resolve; a collection-resident write type is one whose collection that snapshot fills with the tenant's other stored items beside the written one, and there the entry is keyed by NAME, never by an array index — for example `objects.acme_invoice.sharingModel` (an `object` write) or `datasets.acme_invoice_metrics.dimensions[0].field` (a `dataset` write). Every other write type is the sole member of its own collection in that snapshot, so its `[0]` IS the written item and stays positional (for example `flows[0]...`). Which write types are collection-resident is derived, never listed: `@objectstack/lint`'s runtime gate computes the set (`NAME_KEYED_STACK_KEYS`) as the collections it carries as resolution context that a gated write type also lands in, so it changes with the gate, not with this text. An entry whose name cannot be spliced into a dotted path keeps its index. Nested positions inside one named item stay positional (`objects.acme_invoice.indexes[1]`): they index the author's own document. |
| **where** | `string` | ✅ | Human-readable location — `flow "leave_approval" · node "approve"`. Prose for a person; use `path` for anything mechanical. |
| **message** | `string` | ✅ | What is wrong, in the rule author's own words. |
| **hint** | `string` | ✅ | How to fix it. |
| **severity** | `Enum<'error' \| 'warning' \| 'info'>` | ✅ | How the gate treated this finding. `error` means the write was REFUSED (these appear on the 422, never on a 2xx); `warning` / `info` are advisory — the write succeeded and the finding is FYI. |


---

## PublishPackageDraftsRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **message** | `string` | optional | The author's change note (ADR-0146 D1). The package commit records it, and so does every item history row this publish writes, as `change_note` (read back as each history event's `message`). Absent or empty, those rows record the sentence naming the package instead. |
| **backfill** | `{ object: string; field: string }[]` | optional | Opt-in: fields this publish adds whose literal `defaultValue` is written into the object's existing rows once the drafts are live, as the signed-in caller. Absent or `[]`, no row is touched. Each entry names an object and a field and nothing else, and names it once. The whole request is refused `400 VALIDATION_ERROR` before anything is published when an entry is malformed, repeated, or names a field that is not new in this publish or declares no literal default; `backfillApplied` in the response reports the rows written per field. |

### Nested Shape: `PublishPackageDraftsRequest.backfill[number]`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **object** | `string` | ✅ | Name of the object whose existing rows are written. |
| **field** | `string` | ✅ | Name of the field this publish adds to that object. |


---

## PublishPackageDraftsResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **success** | `boolean` | ✅ | True only when every pending draft promoted (`failed` empty) AND at least one item published. A pre-flight refusal or an ADR-0067 D2 rollback answers false on a 200 — but so does a publish with nothing to promote, so false alone is NOT a refusal: read `outcome`, not this boolean or the HTTP status. Always equal to `outcome === 'published'` (pinned). It does NOT cover the best-effort receipts below, each of which reports its own `success`. |
| **outcome** | `Enum<'published' \| 'refused' \| 'nothing_to_publish'>` | ✅ | First-class discriminant for WHICH exit answered — the fact `success` compresses into one boolean. `published`: at least one draft promoted and none refused. `refused`: the batch was refused — a pre-flight violation or the ADR-0067 D2 all-or-nothing rollback; the per-item story is in `failed[]`, which is non-empty exactly on this outcome (the invariant consumers previously had to reverse-engineer, now stated by the producer). `nothing_to_publish`: the package had no pending drafts — nothing landed AND nothing was refused; `success` stays false (a no-op is not a successful publish), which before this field made that answer indistinguishable from a refusal. Producer invariants, pinned in the conformance suites: `success === (outcome === 'published')`; `refused` if and only if `failed.length > 0`; `nothing_to_publish` if and only if `published.length === 0 && failed.length === 0`. Values are lowercase snake, matching the `sys_metadata_audit` outcome vocabulary. |
| **publishedCount** | `integer` | ✅ | Number of drafts promoted to active — `published.length`. 0 on every refusal path (the batch is all-or-nothing, ADR-0067 D2). |
| **failedCount** | `integer` | ✅ | Number of items that did not publish — `failed.length`. On a rollback this counts the WHOLE batch: the causal item plus every sibling marked BATCH_ABORTED. |
| **published** | `{ type: string; name: string; version: string; advisories?: object[] }[]` | ✅ | Every draft promoted to active, in publish order. Empty on every refusal path. |
| **failed** | `{ type: string; name: string; error: string; code?: string; … }[]` | ✅ | Items that did not publish. Because the batch is all-or-nothing (ADR-0067 D2), a non-empty list means NOTHING landed: `published: []`, `publishedCount: 0`. |
| **seedApplied** | `{ success: boolean; inserted?: integer; updated?: integer; error?: string; … }` | optional | Aggregate outcome of materializing EVERY published `seed` body in one multi-pass loader run (cross-seed references need the whole set). Present ONLY when the batch published at least one seed. Two producers, one key: the batch itself self-applies (`applySeedBodies`), and the REST door back-fills the same key for custom protocols that do not — never both (an externalId-less seed would double-insert). Best-effort: a seed problem is surfaced here, never thrown. |
| **materializeApplied** | `{ success: boolean; inserted: integer; updated: integer; failures: object[] }` | optional | ADR-0086 P2 — aggregate result of publish-time materializers across the batch (e.g. `flow`, whose draft credentials become the live ones), including side-effect failures surfaced by the per-item effects loop. Present ONLY when at least one published item had a registered materializer or a side-effect failure. Best-effort, same contract as `seedApplied`. |
| **probes** | `any` | optional | ADR-0038 L3 post-publish runtime probe report — one real read per published artifact (seeded objects have rows, views are readable, widget dataset selections execute). DELIBERATELY OPAQUE in this contract: the key is declared and carried through verbatim, but its inner shape is intentionally not modeled until a consumer needs a field of it. Present only when something was publishable; probes never fail the publish. |
| **commitId** | `string` | optional | ADR-0067 — id of the commit this publish recorded. Absent when nothing published. |
| **backfillApplied** | `{ success: boolean; fields: object[] }` | optional | The opt-in `backfill` request key's receipt: the literal `defaultValue` of each named field this publish added, written into the object's existing rows whose value is absent, page by page, as the caller (permissions, RLS/FLS and record hooks apply as on their own edit). Present ONLY when the request named at least one field and the batch published; absent means nothing was asked. A request that names a field this publish does not add, or one without a literal default, is refused with a 400 before anything is published, so this key never reports a refusal. |
| **unhiddenApps** | `string[]` | optional | ADR-0045 §3 — names of the apps whose `_unpublished` gate this publish cleared (publish = live AND visible; a materialized additive build has no drafts, only this flip). Attached by the REST door, not the protocol helper. Present ONLY when at least one app flipped — on a mid-loop failure it names the apps that DID persist, beside `unhideError` ('s split report). Spelling is a permanent wire contract (see the door's note). |
| **unhideError** | `string` | optional | Present when the ADR-0045 visibility flip failed (wholly or partway): the drafts ARE published, but apps still stored `_unpublished: true` stay externally unobservable. Client-facing text only — undeclared driver text is withheld per ADR-0112; the full cause is in the server log. The route is idempotent: re-run it once the cause is resolved. |
| **rebindError** | `string` | optional | Present when the post-publish `metadata:reloaded` announce failed: everything is published and stored, but boot-cached consumers keep the pre-publish view until re-run or restart (a newly published record-triggered flow does not bind its trigger). Client-facing text only, same ADR-0112 withhold as `unhideError`. |

### Nested Shape: `PublishPackageDraftsResponse.published[number]`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **type** | `string` | ✅ | Metadata type of the promoted draft (canonical singular). |
| **name** | `string` | ✅ | Item name of the promoted draft. |
| **version** | `string` | ✅ | Version token of the promoted row — the same keyed ADR-0008 optimistic-concurrency token the single-item doors return: echo it back as `If-Match` on the next write to this item. A keyed digest of the stored content hash, never the hash itself. Opaque to callers; currently `hmac-sha256:` plus 64 hex characters, but the format is not part of this contract. |
| **advisories** | `{ rule: string; path: string; where: string; message: string; … }[]` | optional | Non-gating findings the runtime authoring gate raised against THIS draft's promotion (the same element shape and the same omitted-when-empty discipline as `PublishMetaItemResponseSchema.advisories`, riding each element rather than a parallel top-level map). Present ONLY when at least one finding was raised — an empty array is never emitted, so an advisory-free batch's response bytes are unchanged. Advisory by construction: every entry is `warning`/`info`, because an `error` finding refuses the promotion and — the batch being all-or-nothing — aborts the whole batch as `failed[]` instead; `failed[]` elements never carry this key. |

### Nested Shape: `PublishPackageDraftsResponse.failed[number]`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **type** | `string` | ✅ | Metadata type of the item that did not publish. |
| **name** | `string` | ✅ | Item name. |
| **error** | `string` | ✅ | What refused it. On a rollback, the causal item carries its real error and every sibling carries the all-or-nothing explanation. A refusal that produced structured findings states a one-sentence HEADLINE here (what failed, where, which rules, how many); the per-path detail rides `issues[]` instead of being restated in this string (consumers rendering both channels were showing every finding twice). |
| **code** | `string` | optional | Machine code for the refusal class (SCREAMING_SNAKE, ADR-0112 vocabulary) — e.g. a pre-flight violation code, or BATCH_ABORTED on the non-causal items of a rolled-back batch. |
| **issues** | `{ rule: string; path: string; where: string; message: string; … }[]` | optional | The structured findings behind the refusal, when the refusing error carried them — today the author-time gate's INVALID_METADATA refusal on the causal item. The producer has emitted this key; declaring it is what lets a typed consumer read it back, and what lets `error` stay a headline without losing the per-path detail. Same element shape as `published[].advisories` and the single-item 422's `issues[]` (— one dialect, declared once). Present ONLY on the causal item and only when the refusal produced structured findings; BATCH_ABORTED siblings never carry it. Absent means "this refusal carried no structured findings", never "no problems". |

### Nested Shape: `PublishPackageDraftsResponse.seedApplied`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **success** | `boolean` | ✅ | False when the seed rows did not fully land. The publish itself still succeeded — check this rather than assuming data went live. |
| **inserted** | `integer` | optional | Rows created by the externalId-keyed upsert. Optional ONLY because the route-level fallback producer (custom protocols that do not self-apply) reports early failures without counters; the in-batch producer always emits both counters. |
| **updated** | `integer` | optional | Rows updated by the externalId-keyed upsert. Same optionality rationale as `inserted`. |
| **error** | `string` | optional | Single failure message, present when the apply failed before the loader ran (including "no readable seed bodies"). When the failure is the seed bodies' own schema refusal, this is a one-sentence headline and the per-path detail rides `issues[]`. |
| **errors** | `any[]` | optional | Per-record failures reported by the seed loader, plus any seed-body read failures. May be present and empty on a clean load. |
| **issues** | `{ path: string; message: string; code?: string }[]` | optional | Structured spec-validation findings behind `error`, present when the apply was refused by the seed bodies' own schema — the declared 422 `seedRequestValidationError` mints. The per-path detail lives HERE, once; `error` stays a one-sentence headline. Absent on non-validation failures (driver faults, unreadable bodies), whose whole story is `error` / `errors[]`. |
| **landed** | `{ object: string; inserted: object; matched: object }[]` | optional | Per object, the stored rows the seed landed on: `inserted` (created by this publish) and `matched` (pre-existing rows a seed key matched), each capped at `SEED_APPLIED_LANDED_IDS_CAP` (200) ids with `total` and `truncated` stating any cut. One entry per object the loader ran, empty lists included. When the load threw part-way (`success: false` with `error`), it lists only the datasets that FINISHED before the throw, and is EMPTY when none had; the dataset that threw is not listed (its rows are in the counts, its object in `error`). ABSENT means "not reported" — the apply failed before any load ran (see `error`), or the producer reports no ids — never "no rows landed". |

### Nested Shape: `PublishPackageDraftsResponse.materializeApplied`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **success** | `boolean` | ✅ | False when any item's materializer failed; the publish still succeeded. |
| **inserted** | `integer` | ✅ | Data-plane rows created across the batch. |
| **updated** | `integer` | ✅ | Data-plane rows updated across the batch. |
| **failures** | `{ type: string; name: string; error: string }[]` | ✅ | Each item whose data-plane projection did NOT land — named per-item, unlike the single-item door's scalar `error`, because one aggregate boolean over N items would hide WHICH ones never went live. |

### Nested Shape: `PublishPackageDraftsResponse.backfillApplied`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **success** | `boolean` | ✅ | False when any field stopped short (its entry carries `error`). The publish itself still succeeded and is not undone. |
| **fields** | `{ object: string; field: string; updated: integer; error?: string; … }[]` | ✅ | One element per requested field, in request order. |


---

## RealtimeConnectRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **transport** | `Enum<'websocket' \| 'sse' \| 'polling'>` | optional | Preferred transport protocol |
| **channels** | `string[]` | optional | Channels to subscribe to on connect |
| **token** | `string` | optional | Authentication token |


---

## RealtimeConnectResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **connectionId** | `string` | ✅ | Unique connection identifier |
| **transport** | `Enum<'websocket' \| 'sse' \| 'polling'>` | ✅ | Negotiated transport protocol |
| **url** | `string` | optional | WebSocket/SSE endpoint URL |


---

## RealtimeDisconnectRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **connectionId** | `string` | optional | Connection ID to disconnect |


---

## RealtimeDisconnectResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **success** | `boolean` | ✅ | Whether disconnection succeeded |


---

## RealtimeSubscribeRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **channel** | `string` | ✅ | Channel name to subscribe to |
| **events** | `string[]` | optional | Specific event types to listen for |
| **filter** | `Record<string, any>` | optional | Event filter criteria |


---

## RealtimeSubscribeResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **subscriptionId** | `string` | ✅ | Unique subscription identifier |
| **channel** | `string` | ✅ | Subscribed channel name |


---

## RealtimeUnsubscribeRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **subscriptionId** | `string` | ✅ | Subscription ID to cancel |


---

## RealtimeUnsubscribeResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **success** | `boolean` | ✅ | Whether unsubscription succeeded |


---

## RegisterDeviceRequest

Register or refresh a push device for the signed-in user

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **token** | `string` | ✅ | Device push notification token |
| **platform** | `Enum<'ios' \| 'android' \| 'web'>` | ✅ | Device platform |
| **deviceId** | `string` | optional | Client-held device id, stable per app install. Re-registering the same deviceId refreshes that device; absent, the server mints one and answers it, and the client keeps it for later registrations and unregister |
| **name** | `string` | optional | Device friendly name |
| **vendor** | `Enum<'apns' \| 'fcm'>` | optional | Push vendor the token belongs to: apns or fcm. Absent, the server derives it from platform: ios means apns, android means fcm. web has no default vendor: a web registration names apns or fcm, and one without vendor is refused |
| **appId** | `string` | optional | App the token was issued to: the iOS bundle ID or the Android application ID |
| **locale** | `string` | optional | Device locale (BCP 47, for example zh-CN) that pushes to this device are rendered in |
| **origin** | `string` | optional | [PLANNED — not yet stored] The origin the Console was loaded from (window.location.origin, for example https://crm.example.com): the address the mobile shell saved for this server. The Console writes it; the server stores it on the device row and every push to the device carries it back, so a shell with several saved servers opens the push on the one that sent it. http or https, a lowercase host and an optional port; a path, another scheme or a bare host is refused |


---

## RegisterDeviceResponse

Answer to a device registration: the device id the client keeps

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **deviceId** | `string` | ✅ | Registered device ID |
| **success** | `boolean` | ✅ | Whether registration succeeded |


---

## RejectAiPendingActionResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **status** | `'rejected'` | ✅ | Always "rejected" |
| **id** | `string` | ✅ | The rejected action id |


---

## RollbackMetaItemResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **success** | `boolean` | ✅ | Whether the rollback landed. |
| **version** | `string` | ✅ | The new live row's ADR-0008 optimistic-concurrency token — the same carrier `saveItem` returns; pass it back as `options.ifMatch`. |
| **seq** | `number` | ✅ | The new live row's history sequence number. |
| **restoredFromVersion** | `number` | ✅ | Which history version was restored. |
| **message** | `string` | optional | Rollback note, when one was recorded. |


---

## RuntimeAuthoringIssue

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **rule** | `string` | ✅ | Stable diagnostic rule id (`flow-multi-write-unfiltered`, `approval-expression-invalid`, …). Machine-readable and stable across releases — the key a renderer groups or suppresses by. |
| **path** | `string` | ✅ | Config path inside the SUBMITTED body (`flows[0].nodes[1].config.multi`), so an editor can jump to the offending key. May be empty when the finding is about the item as a whole. The TOP-LEVEL collection entry follows one rule. The gate evaluates against a private per-write snapshot whose indexes no caller can resolve; a collection-resident write type is one whose collection that snapshot fills with the tenant's other stored items beside the written one, and there the entry is keyed by NAME, never by an array index — for example `objects.acme_invoice.sharingModel` (an `object` write) or `datasets.acme_invoice_metrics.dimensions[0].field` (a `dataset` write). Every other write type is the sole member of its own collection in that snapshot, so its `[0]` IS the written item and stays positional (for example `flows[0]...`). Which write types are collection-resident is derived, never listed: `@objectstack/lint`'s runtime gate computes the set (`NAME_KEYED_STACK_KEYS`) as the collections it carries as resolution context that a gated write type also lands in, so it changes with the gate, not with this text. An entry whose name cannot be spliced into a dotted path keeps its index. Nested positions inside one named item stay positional (`objects.acme_invoice.indexes[1]`): they index the author's own document. |
| **where** | `string` | ✅ | Human-readable location — `flow "leave_approval" · node "approve"`. Prose for a person; use `path` for anything mechanical. |
| **message** | `string` | ✅ | What is wrong, in the rule author's own words. |
| **hint** | `string` | ✅ | How to fix it. |
| **severity** | `Enum<'error' \| 'warning' \| 'info'>` | ✅ | How the gate treated this finding. `error` means the write was REFUSED (these appear on the 422, never on a 2xx); `warning` / `info` are advisory — the write succeeded and the finding is FYI. |


---

## SaveMetaItemRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **type** | `string` | ✅ | Metadata type name |
| **name** | `string` | ✅ | Item name — lowercase snake_case segments, optionally dot-qualified (`crm_lead`, `crm_lead.pipeline`). Slash-compound names are refused at the publish door. |
| **item** | `any` | ✅ | Metadata item definition |
| **parentVersion** | `string \| null` | optional | ADR-0008 optimistic-concurrency pin: the version token the caller believes is current (on the REST door, the `If-Match` request header; the item read's `version` and a save receipt's `version` serve it). Present as a string, a concurrent edit is reported as a 409 conflict instead of silently overwritten. ⚠️ `null` is NOT the same as absent: a present `null` asserts "no current row of this lifecycle" — the first-write pin, refused 409 when a row already exists (on the REST door, `If-None-Match: *`; that header takes `*` alone and never beside `If-Match`, and any other spelling is refused 400) — while an ABSENT key is unpinned: the implementation adopts the current row's hash as the parent (last-write-wins). Nullable because that is the implementation's parameter type, and unlike the reset twin (which folds a present `null` back to the current hash) this verb passes `null` through to the repository's conflict check unchanged. |
| **actor** | `string` | optional | Identity recorded on the write's history event (`recorded_by`, a lookup into `sys_user`) and audit row. On the REST door this is the request's authenticated identity (one producer) — never a caller-supplied header. Absent, the event is recorded actor-less (null), deliberately not attributed to "system". |
| **force** | `boolean` | optional | Destructive-change acknowledgement (`?force=true` on the REST door): skips the safety diff that refuses an `object` save whose body drops fields or narrows types the stored item still carries (409 with the findings otherwise). Only `object` saves reach that diff, so the flag is inert for every other type. Absent = the guard runs. |
| **mode** | `Enum<'draft' \| 'publish'>` | optional | Per-item lifecycle (ADR-0005 drafts): `draft` stages the body as a pending draft overlay (`?mode=draft` on the REST door; the publish door promotes it later); `publish` or ABSENT writes straight to the live `active` row — the legacy default, kept so callers that predate the draft/publish split keep working. Any value other than `draft` is read as `publish`. |
| **packageId** | `string \| null` | optional | ADR-0048 — the software package to bind the saved row to (`sys_metadata.package_id`; `?package=<id>` on the REST door, where `all` and an empty value name no package). Set when authoring inside a Studio package workspace; a named read-only base package is refused. On create the row is stamped with this id; on update an existing binding is preserved, never silently re-bound. Absent or `null` names no package. An `active` save then writes the row of the one package the item's active rows are bound to, as a save naming that package does, is refused `409 RESOURCE_CONFLICT` when they are bound to two or more, and with none writes the env-local overlay (no package stamp); a `draft` save writes the draft of the one package the item's active rows are bound to (a pending change stays in the package it edits), else its package-unbound draft, adopted into that package — the draft the per-item publish naming no package promotes. For an item never published into a package it is the package-unbound draft, else the one package's draft, and with no draft at all the save creates the package-unbound draft. It is refused `409 RESOURCE_CONFLICT` when the active rows (or, with none and no package-unbound draft, the drafts) are bound to two or more packages, as that publish is (see `PublishMetaItemRequestSchema.packageId`). The row written is also the one the parent-version resolution reads: an unpinned save takes its version as the parent (last-write-wins), a pinned one is judged against it, and the item read at the same scope serves it as `version`. |
| **writeFace** | `Enum<'package-duplicate' \| 'meta-envelope' \| 'meta-dispatch' \| 'external-import'>` | optional | Which write door a refusal is being rendered FOR — stated by the SERVER, never by a remote caller: every door builds this request field by field and never spreads a request body into it, so there is no path for a client to smuggle a face in, and a face arriving in a wire body is simply never read. Two refusals branch on it, for different questions: the 409 destructive-change remedy names the acknowledgement mechanism that actually exists on the refusing door (`?force=true` on the REST doors; the dispatcher, the duplicate-package door and the external-table import have none), and the 422 invalid-metadata message adapts to whether a structured `issues[]` channel reaches the consumer beside it. Absent = the conservative default wording. |
| **surface** | `Enum<'copilot' \| 'mcp' \| 'studio' \| 'api'>` | optional | The authoring surface this save is made through (see `MetadataAuthoringSurfaceSchema`). A draft save (`mode: 'draft'`) stamps it as `by` on every key path it changes, the entries the strict draft read serves as `draftChanges`; an active save records no per-key provenance. ABSENT stamps `api`. On the REST door and the runtime dispatcher it is the `?surface=` query parameter, which takes `studio` or `api` and refuses every other value `400 VALIDATION_ERROR`, `copilot` and `mcp` included: `copilot` is stated only by an in-process AI tool calling the save and `mcp` only by the MCP build door after it verified the client's token, never by a remote caller, so an HTTP request cannot mark its own write as the copilot's or an MCP agent's. A value outside the set in-process is refused `400 INVALID_REQUEST`, before anything is written. |
| **client** | `{ id: string; name?: string }` | optional | The OAuth client this save is made by (see `MetadataAuthoringClientSchema`), stated only by the MCP build door, in-process, from the token it verified. Only with `surface: 'mcp'`: a `client` beside any other surface, or a malformed one, is refused `400 INVALID_REQUEST` before anything is written. A draft save stamps it as `client` beside `by: 'mcp'` on every key path it changes; an active save records it nowhere. No HTTP door reads it: the REST door and the dispatcher build this request member by member, so a `client` in a wire body or query is never read. |

### Nested Shape: `SaveMetaItemRequest.client`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **id** | `string` | ✅ | The OAuth client id the verified token was issued to. Opaque — compare it, never parse it. |
| **name** | `string` | optional | The client's registered display name (its OAuth `client_name`) when it registered one, as registered. Absent when it registered none; ⛔ never the id standing in for it. |


---

## SaveMetaItemResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **success** | `boolean` | ✅ |  |
| **version** | `string` | ✅ | Version token of the row this write left, and the token the ADR-0008 optimistic-concurrency chain runs on: send it back as the `If-Match` request header on the next write to that item and a concurrent edit is refused 409 `METADATA_CONFLICT` instead of silently overwritten. The item read serves the same token for the same row (`GetMetaItemResponseSchema.version`). A keyed digest of the stored content hash, never the hash itself (under the registered crypto provider's key, or a process-scoped key while none is registered). Opaque to callers — echo it verbatim, never parse it. Currently emitted as `hmac-sha256:` plus 64 hex characters, but the format is not part of this contract. |
| **seq** | `integer` | ✅ | Monotonic sequence number of the metadata event this write appended to the item history (sys_metadata_history.event_seq). Orders writes; unlike `version` it is not an OCC token. |
| **state** | `Enum<'draft' \| 'active'>` | ✅ | Lifecycle the body was written into: "draft" when the request asked for draft mode (`?mode=draft`), otherwise "active" (published and live). A draft is staged only — it is not served to the runtime until published. |
| **projectionApplied** | `{ success: boolean; error?: string }` | optional | Outcome of the awaited ADR-0094 mutation projector — the post-persist step that materializes this metadata into its derived data-plane read model (e.g. `datasource` → its live connection pool). Present ONLY when a projector is registered for this metadata type, which is why it is optional: its absence means "no projector ran", never "the projection failed". Best-effort by design — a projector failure is reported here and logged, never thrown, so a caller that needs the read model to be live must check `projectionApplied.success` rather than rely on the 200. |
| **advisories** | `{ rule: string; path: string; where: string; message: string; … }[]` | optional | Non-gating findings from the runtime authoring gate — the same shared author-time rules `os validate` / `os build` / `os lint` run, applied to this body on its way to `active`. The write SUCCEEDED; these are what the gate has to say about it anyway (closing D3). Present ONLY when at least one advisory was raised — an empty array is never emitted, so a clean save's response bytes are unchanged and absence means "nothing to report", never "the gate did not run". Advisory by construction: every entry has `severity` `warning` or `info`, because an `error` finding refuses the write and arrives as the 422 `invalid_metadata` envelope instead of here. A caller that ignores this key behaves exactly as before. Runtime-only: the CLI surfaces the same findings on its own stdout, and a Studio / MCP / AI author has no CLI at all, which is the gap this key exists to close. The gate runs on both write doors (D1), and both report: `POST /meta/:type/:name/publish` carries the same key on `PublishMetaItemResponseSchema`. The save also reports here what it did to the row at rest that the author did not write, on a draft save too: one `warning` with rule `withheld-key-dropped` for each stored position it dropped because the type's read redactor withholds it and marks it not carried forward (a connector's undeclared `auth` keys and the undeclared keys of its `retryConfig`, which no door accepts and nothing reads). Its `path` names where the value sat at rest, in the same spelling (`connectors[0].auth.token`). |
| **message** | `string` | optional |  |

### Nested Shape: `SaveMetaItemResponse.projectionApplied`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **success** | `boolean` | ✅ | False when the projector threw; the metadata write itself still succeeded. |
| **error** | `string` | optional | Projector failure message, present only when `success` is false. |

### Nested Shape: `SaveMetaItemResponse.advisories[number]`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **rule** | `string` | ✅ | Stable diagnostic rule id (`flow-multi-write-unfiltered`, `approval-expression-invalid`, …). Machine-readable and stable across releases — the key a renderer groups or suppresses by. |
| **path** | `string` | ✅ | Config path inside the SUBMITTED body (`flows[0].nodes[1].config.multi`), so an editor can jump to the offending key. May be empty when the finding is about the item as a whole. The TOP-LEVEL collection entry follows one rule. The gate evaluates against a private per-write snapshot whose indexes no caller can resolve; a collection-resident write type is one whose collection that snapshot fills with the tenant's other stored items beside the written one, and there the entry is keyed by NAME, never by an array index — for example `objects.acme_invoice.sharingModel` (an `object` write) or `datasets.acme_invoice_metrics.dimensions[0].field` (a `dataset` write). Every other write type is the sole member of its own collection in that snapshot, so its `[0]` IS the written item and stays positional (for example `flows[0]...`). Which write types are collection-resident is derived, never listed: `@objectstack/lint`'s runtime gate computes the set (`NAME_KEYED_STACK_KEYS`) as the collections it carries as resolution context that a gated write type also lands in, so it changes with the gate, not with this text. An entry whose name cannot be spliced into a dotted path keeps its index. Nested positions inside one named item stay positional (`objects.acme_invoice.indexes[1]`): they index the author's own document. |
| **where** | `string` | ✅ | Human-readable location — `flow "leave_approval" · node "approve"`. Prose for a person; use `path` for anything mechanical. |
| **message** | `string` | ✅ | What is wrong, in the rule author's own words. |
| **hint** | `string` | ✅ | How to fix it. |
| **severity** | `Enum<'error' \| 'warning' \| 'info'>` | ✅ | How the gate treated this finding. `error` means the write was REFUSED (these appear on the 422, never on a 2xx); `warning` / `info` are advisory — the write succeeded and the finding is FYI. |


---

## SearchAllHit

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **object** | `string` | ✅ | Name of the object the hit belongs to. |
| **id** | `string` | ✅ | ID of the matched record. |
| **title** | `string` | ✅ | Display title for the hit, resolved in order: the declared primary-title pointer (`nameField`, ADR-0079; deprecated alias `displayNameField` still honored) → the object's deprecated `titleFormat` template → conventional name fields → first/last name → the record ID as a string. |
| **snippet** | `string` | optional | Excerpt cut around the first matched term in a searchable text column, ellipsized at both ends when truncated. ABSENT when no source column literally contains a term (e.g. a pinyin companion match) — absence is a correct answer, not a miss. |
| **record** | `Record<string, any>` | ✅ | The matched record as the engine's find path returns it (row-level security applied, internal fields already stripped). Object-specific — no cross-object field shape is promised beyond "a record of the named object". |


---

## SearchAllPageHit

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **kind** | `'page'` | ✅ | Discriminant, always the literal `page` — kept on the hit (despite the dedicated array) so a client merging record and page hits into one palette list still holds a self-describing element. |
| **name** | `string` | ✅ | Machine name of the page (`page.name`) — the address a client routes to; the same name the metadata read door serves the page under. |
| **title** | `string` | ✅ | Display title: the page `label` resolved through the shared i18n label resolution (exact tag → base language → regional sibling → `default` → `en` → any), falling back to the machine name when no label resolves. |
| **snippet** | `string` | optional | Excerpt cut around the first matched term in the page `description`, ellipsized at both ends when truncated — the same geometry as a record hit's snippet. ABSENT when the match is on the name or label only (the title already shows it) — absence is a correct answer, not a miss. |
| **pageType** | `string` | optional | The page's own declared `type` (`record` \| `app` \| `home` \| …) when the served document carries one — a routing hint, deliberately typed as a plain string so a future page type cannot invalidate an already-produced search response. |


---

## SearchAllResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **query** | `string` | ✅ | The TRIMMED query text the sweep ran with — empty string when the request carried none (the no-scan short-circuit). |
| **hits** | `{ object: string; id: string; title: string; snippet?: string; … }[]` | ✅ | Matched records across objects, in scan order, capped at the overall `limit` (default 20, max 100) with at most `perObject` (default 5, max 25) per object. |
| **pages** | `{ kind: 'page'; name: string; title: string; snippet?: string; … }[]` | ✅ | Published pages whose name, label, or description matches every term (each term may match in a different field; matching folds case), in the served listing's order, capped at `perObject` — the page store is swept as one more container, and more matching pages may exist beyond the cap. Produced ONLY on an unscoped sweep: a request that names `objects` asks for records of those objects and answers `pages: []`. The swept set is exactly what the caller's metadata read door (`GET /api/v1/meta/page`) serves — published state only, never drafts. |
| **totalObjects** | `number` | ✅ | Number of objects the sweep actually SCANNED (searchable, API-enabled, with a resolvable search-field set) — not the number of objects with hits. An object whose table was never provisioned is skipped and not counted. |
| **totalHits** | `number` | ✅ | Number of hits returned — equals `hits.length`. NOT a deployment-wide total-match count: matches beyond `limit` / `perObject` are not counted. |
| **truncated** | `boolean` | ✅ | True when the sweep stopped at the overall `limit` — more matches may exist beyond the returned set. |

### Nested Shape: `SearchAllResponse.hits[number]`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **object** | `string` | ✅ | Name of the object the hit belongs to. |
| **id** | `string` | ✅ | ID of the matched record. |
| **title** | `string` | ✅ | Display title for the hit, resolved in order: the declared primary-title pointer (`nameField`, ADR-0079; deprecated alias `displayNameField` still honored) → the object's deprecated `titleFormat` template → conventional name fields → first/last name → the record ID as a string. |
| **snippet** | `string` | optional | Excerpt cut around the first matched term in a searchable text column, ellipsized at both ends when truncated. ABSENT when no source column literally contains a term (e.g. a pinyin companion match) — absence is a correct answer, not a miss. |
| **record** | `Record<string, any>` | ✅ | The matched record as the engine's find path returns it (row-level security applied, internal fields already stripped). Object-specific — no cross-object field shape is promised beyond "a record of the named object". |

### Nested Shape: `SearchAllResponse.pages[number]`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **kind** | `'page'` | ✅ | Discriminant, always the literal `page` — kept on the hit (despite the dedicated array) so a client merging record and page hits into one palette list still holds a self-describing element. |
| **name** | `string` | ✅ | Machine name of the page (`page.name`) — the address a client routes to; the same name the metadata read door serves the page under. |
| **title** | `string` | ✅ | Display title: the page `label` resolved through the shared i18n label resolution (exact tag → base language → regional sibling → `default` → `en` → any), falling back to the machine name when no label resolves. |
| **snippet** | `string` | optional | Excerpt cut around the first matched term in the page `description`, ellipsized at both ends when truncated — the same geometry as a record hit's snippet. ABSENT when the match is on the name or label only (the title already shows it) — absence is a correct answer, not a miss. |
| **pageType** | `string` | optional | The page's own declared `type` (`record` \| `app` \| `home` \| …) when the served document carries one — a routing hint, deliberately typed as a plain string so a future page type cannot invalidate an already-produced search response. |


---

## SeedAppliedIdList

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **ids** | `string[]` | ✅ | Row ids in the loader's record order (datasets in load order, records as authored), each once — at most `SEED_APPLIED_LANDED_IDS_CAP` (200). |
| **total** | `integer` | ✅ | How many ids the list holds before the cap. Equals `ids.length` unless `truncated`. |
| **truncated** | `boolean` | ✅ | True when the cap cut the list: `total` ids exist and only the first `ids.length` are carried. |


---

## SeedAppliedLanded

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **object** | `string` | ✅ | The object whose rows the lists name. |
| **inserted** | `{ ids: string[]; total: integer; truncated: boolean }` | ✅ | Rows this publish created. The only ones safe to treat as the seed's own. |
| **matched** | `{ ids: string[]; total: integer; truncated: boolean }` | ✅ | Pre-existing rows a seed record's key matched — updated, left as they were, or refused an update. May be a customer's row; never clear these on this receipt's word. |

### Nested Shape: `SeedAppliedLanded.inserted`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **ids** | `string[]` | ✅ | Row ids in the loader's record order (datasets in load order, records as authored), each once — at most `SEED_APPLIED_LANDED_IDS_CAP` (200). |
| **total** | `integer` | ✅ | How many ids the list holds before the cap. Equals `ids.length` unless `truncated`. |
| **truncated** | `boolean` | ✅ | True when the cap cut the list: `total` ids exist and only the first `ids.length` are carried. |

### Nested Shape: `SeedAppliedLanded.matched`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **ids** | `string[]` | ✅ | Row ids in the loader's record order (datasets in load order, records as authored), each once — at most `SEED_APPLIED_LANDED_IDS_CAP` (200). |
| **total** | `integer` | ✅ | How many ids the list holds before the cap. Equals `ids.length` unless `truncated`. |
| **truncated** | `boolean` | ✅ | True when the cap cut the list: `total` ids exist and only the first `ids.length` are carried. |


---

## SetPresenceRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **channel** | `string` | ✅ | Channel to set presence in |
| **state** | `{ userId: string; status: Enum<'online' \| 'away' \| 'busy' \| 'offline'>; lastSeen: string; metadata?: Record<string, any> }` | ✅ | Presence state to set |

### Nested Shape: `SetPresenceRequest.state`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **userId** | `string` | ✅ | User identifier |
| **status** | `Enum<'online' \| 'away' \| 'busy' \| 'offline'>` | ✅ | Current presence status |
| **lastSeen** | `string` | ✅ | ISO 8601 datetime of last activity |
| **metadata** | `Record<string, any>` | optional | Custom presence data (e.g., current page, custom status) |


---

## SetPresenceResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **success** | `boolean` | ✅ | Whether presence was set |


---

## UninstallPackageRequest

Uninstall package request

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **id** | `string` | ✅ | Package ID to uninstall |


---

## UninstallPackageResponse

Uninstall package response

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **id** | `string` | ✅ | Uninstalled package ID |
| **success** | `boolean` | ✅ | Whether uninstall succeeded |
| **message** | `string` | optional | Uninstall status message |


---

## UnregisterDeviceRequest

Unregister one of the signed-in user's push devices

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **deviceId** | `string` | ✅ | Device ID to unregister |


---

## UnregisterDeviceResponse

Answer to a device unregistration

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **success** | `boolean` | ✅ | Whether unregistration succeeded |


---

## UpdateAiConversationRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **title** | `string` | optional | New title |
| **metadata** | `Record<string, any>` | optional | New metadata |


---

## UpdateDataRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **object** | `string` | ✅ | The object name. |
| **id** | `string` | ✅ | The ID of the record to update. |
| **data** | `Record<string, any>` | ✅ | The fields to update (partial update). |
| **expectedVersion** | `string` | optional | Optimistic concurrency token (typically the `updated_at` value the client read). When provided, the server compares it against the current record version and returns 409 CONCURRENT_UPDATE if they differ. Optional — omit to skip the check. The quoted-empty entity-tag (`""`) is refused 400 VALIDATION_FAILED, not treated as omitted. |


---

## UpdateDataResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **object** | `string` | ✅ | Object name |
| **id** | `string` | ✅ | Updated record ID |
| **record** | `Record<string, any>` | ✅ | Updated record |
| **droppedFields** | `{ object: string; fields: string[]; reason: Enum<'readonly' \| 'readonly_when' \| 'primary_key' \| 'computed' \| 'stored_expression'> }[]` | optional | Write-observability: caller-supplied fields the engine LEGALLY stripped from the write before persisting — static `readonly` or a TRUE `readonlyWhen` predicate. Present ONLY when ≥1 field was dropped; the update still succeeded without them (status/success semantics unchanged — stripping is legitimate, not an error). REST additionally surfaces this as the `X-ObjectStack-Dropped-Fields` response header. Optional — omit-when-empty keeps the shape backward-compatible for existing clients that only read `record`. |
| **warnings** | `{ field: string; code: string; message: string; rule?: string; … }[]` | optional | The advisory validation-rule hits of THIS write: each `severity: 'warning'` or `'info'` rule whose verdict on the written record was "violated", with its `rule` name, `severity`, `field`, `code` and author-written `message` — the same element the validate-only preview reports in a row's `warnings`. Advisory rules never block, so the write succeeded; this is the advisory to show the person who made it. Present ONLY when at least one advisory rule hit; a write with no hit omits the key. Hits of a nested write a hook or a flow made inside this one are not included, and a rule that could not be evaluated is not a hit. |
| **withheldMentions** | `string[]` | optional | The ids of the users this write @mentioned whose mention notification was WITHHELD by the mention recipient gate, so they will not be notified: the gate's own verdict on THIS write, ids only, never the reason. Only a `sys_comment` write fills it, and only on the requester's own answer to the request that made the write; on an update, only a mention the update added is judged. The comment was saved with every mention; the requester is told here that the named users will not be notified. Response-only: never stored on the comment (a stored member would tell every reader who cannot read the thread). Present ONLY when at least one mention was withheld; otherwise the key is omitted. |
| **notice** | `string` | optional | One author-written message telling the person who made THIS write what the save set in motion (for example "Sent to your manager for approval"), set by a hook running on the write through the write's own `onNotice` listener (`WriteObservabilityOptions`; a hook body calls it as `ctx.notice(message)`), or by a `save_notice` node of a flow the write triggered. Info severity: it reports an outcome, never a problem, so it is distinct from `warnings` (the advisory validation-rule hits) and a client shows it with, or in place of, its own success message. One message: when more than one call sets it, the LAST one wins. Present ONLY when a hook set a non-empty message on a write that succeeded; otherwise the key is omitted. A nested write a hook or a flow made inside this one carries its own options, so its notice never reaches this answer. Response-only: never stored on the record. |

### Nested Shape: `UpdateDataResponse.droppedFields[number]`

A write-path strip event: caller-supplied fields legally dropped from the payload, or written with the engine's computed value in place of the caller's

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **object** | `string` | ✅ | Object the write targeted (resolved object name) |
| **fields** | `string[]` | ✅ | Caller-supplied field names whose value the engine did not store: removed from the write payload, or (`stored_expression`) written with the computed value in place of the caller's |
| **reason** | `Enum<'readonly' \| 'readonly_when' \| 'primary_key' \| 'computed' \| 'stored_expression'>` | ✅ | Why the fields were dropped: static readonly, a TRUE readonlyWhen predicate, the primary-key strip of a payload id the engine ruled is not an identifier, a computed (formula) field no driver has a column for, or a stored field whose declared expression the engine computes on every write, so the caller's different value was replaced by the computed one |

### Nested Shape: `UpdateDataResponse.warnings[number]`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **field** | `string` | ✅ | The field the finding is about (`_record` for an object-level rule). |
| **code** | `string` | ✅ | Machine-readable finding code, e.g. `required`, `invalid_type`, `rule_violation`. |
| **message** | `string` | ✅ | Human-readable message — a validation rule's author-written text where one exists. |
| **rule** | `string` | optional | The `name` of the ADVISORY validation rule this finding reports. Present only on an advisory rule hit. |
| **severity** | `Enum<'warning' \| 'info'>` | optional | The advisory rule's declared severity, `warning` or `info`. Present only on an advisory rule hit. |


---

## UpdateManyDataRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **records** | `{ id: string; data: Record<string, any> }[]` | ✅ | Array of records to update (server caps the count — see batch.maxBatchSize) |
| **options** | `{ atomic?: boolean; returnRecords?: boolean; continueOnError?: boolean }` | optional | Update options |
| **object** | `string` | ✅ | Object name |

### Nested Shape: `UpdateManyDataRequest.records[number]`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **id** | `string` | ✅ | Record ID |
| **data** | `Record<string, any>` | ✅ | Fields to update |

### Nested Shape: `UpdateManyDataRequest.options`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **atomic** | `boolean` | optional (default: `false`) | Opt-in all-or-nothing. When explicitly true the whole batch runs inside ONE engine transaction: the first failure rolls back every prior write, and the response reports zero successes — each row carries `errors[0].code` ROLLED_BACK (written, then undone), the causal row its own error, and rows never reached NOT_ATTEMPTED. A runtime that cannot roll back REFUSES the request (501 NOT_IMPLEMENTED) rather than silently degrading to best-effort — probe `capabilities.transactionalBatch` on /discovery first. Takes precedence over continueOnError. Default false: sequential best-effort. |
| **returnRecords** | `boolean` | optional (default: `false`) | If true, return full record data in response |
| **continueOnError** | `boolean` | optional (default: `false`) | If true (and atomic=false), continue processing remaining records after errors. Default false: the first failure ENDS the run — records before it stay written (nothing is rolled back on this arm), and every record after it is reported `errors[0].code` NOT_ATTEMPTED rather than omitted, so `results` always covers all `total` records and `succeeded + failed === total`. |
| **validateOnly** | `never` | optional | [REMOVED] `options.validateOnly` was removed from BatchOptions in @objectstack/spec 17.0.0. It was never implemented: the batch surfaces persisted regardless, so a "dry-run" would have silently executed. There is no dry-run today — drop the key. If you need to preview a batch without writing, open an issue so it can be designed (no-commit cascade / constraint semantics) and reintroduced as a flag that actually holds. |


---

## UpdateManyDataResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **success** | `boolean` | ✅ | Operation success status |
| **error** | `{ code: Enum<'VALIDATION_ERROR' \| 'INVALID_FIELD' \| 'MISSING_REQUIRED_FIELD' \| …>; declaredCode?: string; message: string; userMessage?: string; … }` | optional | Error details if success is false |
| **meta** | `{ timestamp: string; duration?: integer; requestId?: string; traceId?: string }` | optional | Response metadata |
| **operation** | `Enum<'create' \| 'update' \| 'upsert' \| 'delete'>` | optional | Operation type that was performed |
| **total** | `number` | ✅ | Total number of records in the batch |
| **succeeded** | `number` | ✅ | Number of records that succeeded |
| **failed** | `number` | ✅ | Number of records that failed |
| **results** | `{ id?: string; success: boolean; errors?: object[]; data?: Record<string, any>; … }[]` | ✅ | Detailed results for each record |

### Nested Shape: `UpdateManyDataResponse.error`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **code** | `Enum<'VALIDATION_ERROR' \| 'INVALID_FIELD' \| 'MISSING_REQUIRED_FIELD' \| …>` | ✅ | Error code (e.g. VALIDATION_ERROR; StandardErrorCode ∪ the ledger the serving side registers — ERROR_CODE_LEDGER for framework packages) |
| **declaredCode** | `string` | optional | The producer-declared code, verbatim, when it is not a member of the closed `code` vocabulary — the open, author-authored channel (app-specific spellings; ADR-0112) |
| **message** | `string` | ✅ | Readable error message |
| **userMessage** | `string` | optional | Producer-marked user-facing refusal text, verbatim. Present exactly when the producer opted in at throw time; consumers render it to end users and keep their generic substitution for anything unmarked. Status-agnostic; never replaces `message`. |
| **refusal** | `true` | optional | Producer-declared: the 5xx this envelope carries is a deliberate refusal whose `message` is authored for the caller, so a boundary that reads the declaration keeps it verbatim (until the withhold arms read it, a declared refusal is still withheld). Absent (the default) on a declared fault, whose `message` is withheld from the body and logged for the operator; redundant on a 4xx. Presence is the declaration — `true` is the only value. |
| **category** | `string` | optional | Error category (e.g. validation, authorization) |
| **httpStatus** | `integer` | optional | HTTP status of the response carrying this error |
| **details** | `any` | optional | Additional error context (e.g. field validation errors) |
| **requestId** | `string` | optional | Request ID for tracking |

### Nested Shape: `UpdateManyDataResponse.meta`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **timestamp** | `string` | ✅ |  |
| **duration** | `integer` | optional | Server-side processing duration in milliseconds |
| **requestId** | `string` | optional |  |
| **traceId** | `string` | optional |  |

### Nested Shape: `UpdateManyDataResponse.results[number]`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **id** | `string` | optional | Record ID if operation succeeded |
| **success** | `boolean` | ✅ | Whether this record was processed successfully |
| **errors** | `{ code: Enum<'VALIDATION_ERROR' \| 'INVALID_FIELD' \| 'MISSING_REQUIRED_FIELD' \| …>; declaredCode?: string; message: string; userMessage?: string; … }[]` | optional | Array of errors if operation failed. Branch on `errors[0].code` — an atomic batch that rolled back marks rows that were written then undone with code ROLLED_BACK and rows never reached with NOT_ATTEMPTED, while the causal row keeps its own error. A NON-atomic batch that stopped (the `continueOnError: false` default) marks its un-attempted tail with the same NOT_ATTEMPTED code — rows before the failure stay written and keep reporting success, since nothing was rolled back. |
| **data** | `Record<string, any>` | optional | Full record data (if returnRecords=true) |
| **index** | `number` | optional | Index of the record in the request array |
| **droppedFields** | `{ object: string; fields: string[]; reason: Enum<'readonly' \| 'readonly_when' \| 'primary_key' \| 'computed' \| 'stored_expression'> }[]` | optional | Write-observability: caller-supplied fields LEGALLY stripped from THIS row before it was written — static `readonly` / TRUE `readonlyWhen` on update, or the in-engine static `readonly` strip on create. Per-row because a batch can drop different fields on different rows (`readonlyWhen` is record-state-dependent). Present ONLY when ≥1 field was dropped for this row; the row still succeeded (success unchanged). A single response header cannot say which row dropped what, so this body field stays the canonical per-row channel; REST also sends the `X-ObjectStack-Dropped-Fields` response header on the answer, naming the union of the rows' drops (each field and reason once). Optional — omit-when-empty keeps the shape backward-compatible. |


---

## UpdateNotificationPreferencesRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **preferences** | `{ email?: boolean; push?: boolean; inApp?: boolean; digest?: Enum<'none' \| 'daily' \| 'weekly'>; … }` | ✅ | Preferences to update |

### Nested Shape: `UpdateNotificationPreferencesRequest.preferences`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **email** | `boolean` | optional (default: `true`) | Receive email notifications |
| **push** | `boolean` | optional (default: `true`) | Receive push notifications |
| **inApp** | `boolean` | optional (default: `true`) | Receive in-app notifications |
| **digest** | `Enum<'none' \| 'daily' \| 'weekly'>` | optional (default: `"none"`) | Email digest frequency |
| **channels** | `Record<string, { enabled?: boolean; email?: boolean; push?: boolean }>` | optional | Per-channel notification preferences |


---

## UpdateNotificationPreferencesResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **preferences** | `{ email?: boolean; push?: boolean; inApp?: boolean; digest?: Enum<'none' \| 'daily' \| 'weekly'>; … }` | ✅ | Updated notification preferences |

### Nested Shape: `UpdateNotificationPreferencesResponse.preferences`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **email** | `boolean` | optional (default: `true`) | Receive email notifications |
| **push** | `boolean` | optional (default: `true`) | Receive push notifications |
| **inApp** | `boolean` | optional (default: `true`) | Receive in-app notifications |
| **digest** | `Enum<'none' \| 'daily' \| 'weekly'>` | optional (default: `"none"`) | Email digest frequency |
| **channels** | `Record<string, { enabled?: boolean; email?: boolean; push?: boolean }>` | optional | Per-channel notification preferences |


---

## ValidateDataIssue

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **field** | `string` | ✅ | The field the finding is about (`_record` for an object-level rule). |
| **code** | `string` | ✅ | Machine-readable finding code, e.g. `required`, `invalid_type`, `rule_violation`. |
| **message** | `string` | ✅ | Human-readable message — a validation rule's author-written text where one exists. |
| **rule** | `string` | optional | The `name` of the ADVISORY validation rule this finding reports. Present only on an advisory rule hit. |
| **severity** | `Enum<'warning' \| 'info'>` | optional | The advisory rule's declared severity, `warning` or `info`. Present only on an advisory rule hit. |


---

## ValidateDataRequest

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **object** | `string` | ✅ | The object name. |
| **data** | `Record<string, any> \| Record<string, any>[]` | ✅ | A candidate record, or an array of them. Nothing is persisted. |
| **mode** | `Enum<'insert' \| 'update'>` | optional | Which write the verdict should predict. `insert` (default) walks every declared field, so a missing required field is a finding; `update` checks only the supplied keys, matching a PATCH. An `update` row that carries its `id` is judged as the by-id update judges it: its rules read the stored row that id names, merged with the supplied keys. The stored row is read with the caller's own read access, so a row the caller cannot read is judged on the supplied keys alone, as a row with no `id` is. |


---

## ValidateDataResponse

### Properties

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **object** | `string` | ✅ | The object name. |
| **mode** | `Enum<'insert' \| 'update'>` | ✅ | The write mode the verdict was reached for. |
| **valid** | `boolean` | ✅ | True when EVERY row is valid — the whole-set answer. |
| **results** | `{ valid: boolean; errors: object[]; warnings: object[]; droppedFields?: object[] }[]` | ✅ | Per-row verdicts, in submission order. |
| **posture** | `{ valueShapeStrict: boolean; mediaValueShapeStrict: boolean }` | ✅ | The ADR-0104 posture the verdict was reached under — reported because it is the difference between "this row is fine" and "this row is fine HERE". The same row can be an error on a self-certified deployment and an admitted warning on an un-migrated one, and a caller explaining a verdict needs to know which it got. An unconditionally-strict preview was considered and rejected (option B): it would fail rows on every un-migrated deployment that the write would have accepted. |

### Nested Shape: `ValidateDataResponse.results[number]`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **valid** | `boolean` | ✅ | True when this row would be accepted by the write path. |
| **errors** | `{ field: string; code: string; message: string; rule?: string; … }[]` | ✅ | Findings that would REJECT this row. Empty when valid. |
| **warnings** | `{ field: string; code: string; message: string; rule?: string; … }[]` | ✅ | Findings the target deployment ADMITS rather than rejects; the row is valid, and the write would store it. Two populations, in this order: ADR-0104 value shapes admitted under a warn-first posture (the write logs the same complaint), then the hits of ADVISORY validation rules (`severity: 'warning'` or `'info'`), each naming its `rule` and `severity` — the same elements the write answers as its own `warnings`. An advisory hit is reported only on a row the verdict accepts, as on the write, which answers a refused row with its error. |
| **droppedFields** | `{ object: string; fields: string[]; reason: Enum<'readonly' \| 'readonly_when' \| 'primary_key' \| 'computed' \| 'stored_expression'> }[]` | optional | Write-observability for the preview: caller-supplied fields the write would LEGALLY strip from THIS row, one event per reason, in the shape and reason vocabulary the write itself reports. A strip is not a finding: `valid` is unaffected, and the fields appear in neither `errors` nor `warnings`. Present only when at least one field would be dropped. A server that does not produce this report omits the key too, so an absent key alone does not prove nothing would be dropped. |

### Nested Shape: `ValidateDataResponse.posture`

| Property | Type | Required | Description |
| :--- | :--- | :--- | :--- |
| **valueShapeStrict** | `boolean` | ✅ | True when this deployment rejects non-conforming value shapes (ADR-0104 self-certified). |
| **mediaValueShapeStrict** | `boolean` | ✅ | The same, for media field value shapes. |


---