@knowledge-stack/ksapi 1.131.5 → 1.133.0

package/.openapi-generator/FILES

@@ -79,7 +79,7 @@ docs/CreateThreadMessageRequest.md
 docs/CreateThreadRequest.md
 docs/CreateWorkflowDefinitionRequest.md
 docs/DataSourceCatalogResponse.md
-docs/DataSourceDescriptionResponse.md
+docs/DataSourceDescribeResponse.md
 docs/DataSourceDetailResponse.md
 docs/DataSourceEngine.md
 docs/DataSourceQueryRequest.md
@@ -88,6 +88,7 @@ docs/DataSourceResponse.md
 docs/DataSourceSchemaListItem.md
 docs/DataSourceSchemaListResponse.md
 docs/DataSourceSchemaResponse.md
+docs/DataSourceSyncResponse.md
 docs/DataSourceTableResponse.md
 docs/DataSourcesApi.md
 docs/DefaultApi.md
@@ -245,9 +246,13 @@ docs/ResolvedReferenceOutput.md
 docs/ResponseSendPwResetEmail.md
 docs/RevisionEdit.md
 docs/RootResponse.md
+docs/RunFolder.md
+docs/RunFolderRef.md
 docs/SSOInitiateResponse.md
 docs/ScoredChunkResponse.md
 docs/SearchSortOrder.md
+docs/SearchTablesRequest.md
+docs/SearchTablesResponse.md
 docs/SearchType.md
 docs/SearchablePartType.md
 docs/SeatQuotaStatus.md
@@ -276,7 +281,9 @@ docs/SubtreeChunkGroup.md
 docs/SubtreeChunksResponse.md
 docs/SupportedIdP.md
 docs/SupportedLanguage.md
+docs/SystemJobsApi.md
 docs/TableCellChange.md
+docs/TableSearchResult.md
 docs/TagOrder.md
 docs/TagResponse.md
 docs/TagsApi.md
@@ -293,6 +300,8 @@ docs/TenantUserOrder.md
 docs/TenantUserResponse.md
 docs/TenantUserRole.md
 docs/TenantsApi.md
+docs/TestConnectionRequest.md
+docs/TestConnectionResponse.md
 docs/TextDiff.md
 docs/TextPart.md
 docs/TextPartOrReasoningPartOrToolPartOrDocEditPart.md
@@ -303,6 +312,7 @@ docs/ThreadMessagesApi.md
 docs/ThreadOrder.md
 docs/ThreadResponse.md
 docs/ThreadsApi.md
+docs/ToolDisplayType.md
 docs/ToolPart.md
 docs/ToolStatus.md
 docs/TransferOwnerRequest.md
@@ -357,7 +367,6 @@ docs/WorkflowRunSnapshot.md
 docs/WorkflowRunSummaryResponse.md
 docs/WorkflowRunsApi.md
 docs/WorkflowSummaryResponse.md
-docs/WorkflowsApi.md
 docs/XlsxCellAnchorInput.md
 docs/XlsxCellAnchorInputOrDocxParagraphAnchorInput.md
 docs/XlsxCellAnchorOutput.md
@@ -386,6 +395,7 @@ src/apis/PathPartsApi.ts
 src/apis/PublicApi.ts
 src/apis/SectionsApi.ts
 src/apis/SubscriptionsApi.ts
+src/apis/SystemJobsApi.ts
 src/apis/TagsApi.ts
 src/apis/TenantGroupsApi.ts
 src/apis/TenantsApi.ts
@@ -397,7 +407,6 @@ src/apis/UsersApi.ts
 src/apis/WorkflowDefinitionsApi.ts
 src/apis/WorkflowMemoryApi.ts
 src/apis/WorkflowRunsApi.ts
-src/apis/WorkflowsApi.ts
 src/apis/index.ts
 src/index.ts
 src/models/AcceptDisclaimerRequest.ts
@@ -469,7 +478,7 @@ src/models/CreateThreadMessageRequest.ts
 src/models/CreateThreadRequest.ts
 src/models/CreateWorkflowDefinitionRequest.ts
 src/models/DataSourceCatalogResponse.ts
-src/models/DataSourceDescriptionResponse.ts
+src/models/DataSourceDescribeResponse.ts
 src/models/DataSourceDetailResponse.ts
 src/models/DataSourceEngine.ts
 src/models/DataSourceQueryRequest.ts
@@ -478,6 +487,7 @@ src/models/DataSourceResponse.ts
 src/models/DataSourceSchemaListItem.ts
 src/models/DataSourceSchemaListResponse.ts
 src/models/DataSourceSchemaResponse.ts
+src/models/DataSourceSyncResponse.ts
 src/models/DataSourceTableResponse.ts
 src/models/DiffFormat.ts
 src/models/DiffRow.ts
@@ -622,9 +632,13 @@ src/models/ResolvedReferenceOutput.ts
 src/models/ResponseSendPwResetEmail.ts
 src/models/RevisionEdit.ts
 src/models/RootResponse.ts
+src/models/RunFolder.ts
+src/models/RunFolderRef.ts
 src/models/SSOInitiateResponse.ts
 src/models/ScoredChunkResponse.ts
 src/models/SearchSortOrder.ts
+src/models/SearchTablesRequest.ts
+src/models/SearchTablesResponse.ts
 src/models/SearchType.ts
 src/models/SearchablePartType.ts
 src/models/SeatQuotaStatus.ts
@@ -652,6 +666,7 @@ src/models/SubtreeChunksResponse.ts
 src/models/SupportedIdP.ts
 src/models/SupportedLanguage.ts
 src/models/TableCellChange.ts
+src/models/TableSearchResult.ts
 src/models/TagOrder.ts
 src/models/TagResponse.ts
 src/models/TemporalWorkflowStatusResponse.ts
@@ -665,6 +680,8 @@ src/models/TenantUserEditRequest.ts
 src/models/TenantUserOrder.ts
 src/models/TenantUserResponse.ts
 src/models/TenantUserRole.ts
+src/models/TestConnectionRequest.ts
+src/models/TestConnectionResponse.ts
 src/models/TextDiff.ts
 src/models/TextPart.ts
 src/models/TextPartOrReasoningPartOrToolPartOrDocEditPart.ts
@@ -673,6 +690,7 @@ src/models/ThreadMessageDetails.ts
 src/models/ThreadMessageResponse.ts
 src/models/ThreadOrder.ts
 src/models/ThreadResponse.ts
+src/models/ToolDisplayType.ts
 src/models/ToolPart.ts
 src/models/ToolStatus.ts
 src/models/TransferOwnerRequest.ts

package/README.md

@@ -1,4 +1,4 @@
-# @knowledge-stack/ksapi@1.131.5
+# @knowledge-stack/ksapi@1.133.0
 
 A TypeScript SDK client for the localhost API.
 
@@ -105,14 +105,17 @@ All URIs are relative to *http://localhost:8000*
 *DataSourcesApi* | [**deleteDataSource**](docs/DataSourcesApi.md#deletedatasource) | **DELETE** /v1/data-sources/{data_source_id} | Delete Data Source Handler
 *DataSourcesApi* | [**deleteDataSourceSchema**](docs/DataSourcesApi.md#deletedatasourceschema) | **DELETE** /v1/data-sources/{data_source_id}/schemas/{schema_id} | Delete Data Source Schema Handler
 *DataSourcesApi* | [**deleteDataSourceTable**](docs/DataSourcesApi.md#deletedatasourcetable) | **DELETE** /v1/data-sources/{data_source_id}/tables/{table_id} | Delete Data Source Table Handler
-*DataSourcesApi* | [**generateDataSourceDescription**](docs/DataSourcesApi.md#generatedatasourcedescription) | **POST** /v1/data-sources/{data_source_id}/describe | Generate Data Source Description Handler
+*DataSourcesApi* | [**describeDataSourceTables**](docs/DataSourcesApi.md#describedatasourcetables) | **POST** /v1/data-sources/{data_source_id}/describe | Describe Data Source Tables Handler
 *DataSourcesApi* | [**getDataSource**](docs/DataSourcesApi.md#getdatasource) | **GET** /v1/data-sources/{data_source_id} | Get Data Source Handler
 *DataSourcesApi* | [**getDataSourceCatalog**](docs/DataSourcesApi.md#getdatasourcecatalog) | **GET** /v1/data-sources/{data_source_id}/catalog | Get Data Source Catalog Handler
 *DataSourcesApi* | [**listDataSourceSchemas**](docs/DataSourcesApi.md#listdatasourceschemas) | **GET** /v1/data-sources/{data_source_id}/schemas | List Data Source Schemas Handler
 *DataSourcesApi* | [**modelDataSourceTable**](docs/DataSourcesApi.md#modeldatasourcetable) | **POST** /v1/data-sources/{data_source_id}/tables | Model Data Source Table Handler
 *DataSourcesApi* | [**modelDataSourceTables**](docs/DataSourcesApi.md#modeldatasourcetables) | **POST** /v1/data-sources/{data_source_id}/tables/batch | Model Data Source Tables Handler
 *DataSourcesApi* | [**queryDataSource**](docs/DataSourcesApi.md#querydatasource) | **POST** /v1/data-sources/{data_source_id}/query | Query Data Source Handler
+*DataSourcesApi* | [**searchDataSourceTables**](docs/DataSourcesApi.md#searchdatasourcetables) | **POST** /v1/data-sources/tables/search | Search Data Source Tables Handler
+*DataSourcesApi* | [**syncDataSource**](docs/DataSourcesApi.md#syncdatasource) | **POST** /v1/data-sources/{data_source_id}/sync | Sync Data Source Handler
 *DataSourcesApi* | [**testDataSourceConnection**](docs/DataSourcesApi.md#testdatasourceconnection) | **POST** /v1/data-sources/{data_source_id}/test | Test Data Source Connection Handler
+*DataSourcesApi* | [**testDataSourceConnectionFresh**](docs/DataSourcesApi.md#testdatasourceconnectionfresh) | **POST** /v1/data-sources/test-connection | Test Data Source Connection Fresh Handler
 *DataSourcesApi* | [**updateDataSource**](docs/DataSourcesApi.md#updatedatasourceoperation) | **PATCH** /v1/data-sources/{data_source_id} | Update Data Source Handler
 *DataSourcesApi* | [**updateDataSourceTable**](docs/DataSourcesApi.md#updatedatasourcetable) | **PATCH** /v1/data-sources/{data_source_id}/tables/{table_id} | Update Data Source Table Handler
 *DefaultApi* | [**healthCheck**](docs/DefaultApi.md#healthcheck) | **GET** /healthz | Health Check Handler
@@ -182,6 +185,11 @@ All URIs are relative to *http://localhost:8000*
 *SectionsApi* | [**updateSection**](docs/SectionsApi.md#updatesectionoperation) | **PATCH** /v1/sections/{section_id} | Update Section Handler
 *SubscriptionsApi* | [**changeTenantSubscription**](docs/SubscriptionsApi.md#changetenantsubscription) | **POST** /v1/tenants/{tenant_id}/subscriptions | Change Tenant Subscription Handler
 *SubscriptionsApi* | [**getTenantSubscription**](docs/SubscriptionsApi.md#gettenantsubscription) | **GET** /v1/tenants/{tenant_id}/subscriptions | Get Tenant Subscription Handler
+*SystemJobsApi* | [**cancelTemporalWorkflow**](docs/SystemJobsApi.md#canceltemporalworkflow) | **DELETE** /v1/system-jobs/{workflow_id} | Cancel Temporal Workflow Handler
+*SystemJobsApi* | [**dvWorkflowRerun**](docs/SystemJobsApi.md#dvworkflowrerun) | **POST** /v1/system-jobs/document_versions/{workflow_id} | Dv Workflow Rerun Handler
+*SystemJobsApi* | [**getDvWorkflow**](docs/SystemJobsApi.md#getdvworkflow) | **GET** /v1/system-jobs/document_versions/{workflow_id} | Get Dv Workflow Handler
+*SystemJobsApi* | [**getTemporalWorkflowStatus**](docs/SystemJobsApi.md#gettemporalworkflowstatus) | **GET** /v1/system-jobs/{workflow_id} | Get Temporal Workflow Status Handler
+*SystemJobsApi* | [**listDvWorkflows**](docs/SystemJobsApi.md#listdvworkflows) | **GET** /v1/system-jobs/document_versions | List Dv Workflows Handler
 *TagsApi* | [**createTag**](docs/TagsApi.md#createtagoperation) | **POST** /v1/tags | Create Tag Handler
 *TagsApi* | [**deleteTag**](docs/TagsApi.md#deletetag) | **DELETE** /v1/tags/{tag_id} | Delete Tag Handler
 *TagsApi* | [**getTag**](docs/TagsApi.md#gettag) | **GET** /v1/tags/{tag_id} | Get Tag Handler
@@ -258,11 +266,6 @@ All URIs are relative to *http://localhost:8000*
 *WorkflowRunsApi* | [**stopWorkflowRun**](docs/WorkflowRunsApi.md#stopworkflowrun) | **POST** /v1/workflow-runs/{run_id}/stop | Stop Workflow Run Handler
 *WorkflowRunsApi* | [**updateWorkflowRun**](docs/WorkflowRunsApi.md#updateworkflowrunoperation) | **PATCH** /v1/workflow-runs/{run_id} | Update Workflow Run Handler
 *WorkflowRunsApi* | [**workflowRunCallback**](docs/WorkflowRunsApi.md#workflowruncallbackoperation) | **POST** /v1/workflow-runs/{run_id}/callback | Workflow Run Callback Handler
-*WorkflowsApi* | [**cancelTemporalWorkflow**](docs/WorkflowsApi.md#canceltemporalworkflow) | **DELETE** /v1/workflows/{workflow_id} | Cancel Temporal Workflow Handler
-*WorkflowsApi* | [**dvWorkflowRerun**](docs/WorkflowsApi.md#dvworkflowrerun) | **POST** /v1/workflows/document_versions/{workflow_id} | Dv Workflow Rerun Handler
-*WorkflowsApi* | [**getDvWorkflow**](docs/WorkflowsApi.md#getdvworkflow) | **GET** /v1/workflows/document_versions/{workflow_id} | Get Dv Workflow Handler
-*WorkflowsApi* | [**getTemporalWorkflowStatus**](docs/WorkflowsApi.md#gettemporalworkflowstatus) | **GET** /v1/workflows/{workflow_id} | Get Temporal Workflow Status Handler
-*WorkflowsApi* | [**listDvWorkflows**](docs/WorkflowsApi.md#listdvworkflows) | **GET** /v1/workflows/document_versions | List Dv Workflows Handler
 
 
 ### Models
@@ -336,7 +339,7 @@ All URIs are relative to *http://localhost:8000*
 - [CreateThreadRequest](docs/CreateThreadRequest.md)
 - [CreateWorkflowDefinitionRequest](docs/CreateWorkflowDefinitionRequest.md)
 - [DataSourceCatalogResponse](docs/DataSourceCatalogResponse.md)
-- [DataSourceDescriptionResponse](docs/DataSourceDescriptionResponse.md)
+- [DataSourceDescribeResponse](docs/DataSourceDescribeResponse.md)
 - [DataSourceDetailResponse](docs/DataSourceDetailResponse.md)
 - [DataSourceEngine](docs/DataSourceEngine.md)
 - [DataSourceQueryRequest](docs/DataSourceQueryRequest.md)
@@ -345,6 +348,7 @@ All URIs are relative to *http://localhost:8000*
 - [DataSourceSchemaListItem](docs/DataSourceSchemaListItem.md)
 - [DataSourceSchemaListResponse](docs/DataSourceSchemaListResponse.md)
 - [DataSourceSchemaResponse](docs/DataSourceSchemaResponse.md)
+- [DataSourceSyncResponse](docs/DataSourceSyncResponse.md)
 - [DataSourceTableResponse](docs/DataSourceTableResponse.md)
 - [DiffFormat](docs/DiffFormat.md)
 - [DiffRow](docs/DiffRow.md)
@@ -489,9 +493,13 @@ All URIs are relative to *http://localhost:8000*
 - [ResponseSendPwResetEmail](docs/ResponseSendPwResetEmail.md)
 - [RevisionEdit](docs/RevisionEdit.md)
 - [RootResponse](docs/RootResponse.md)
+- [RunFolder](docs/RunFolder.md)
+- [RunFolderRef](docs/RunFolderRef.md)
 - [SSOInitiateResponse](docs/SSOInitiateResponse.md)
 - [ScoredChunkResponse](docs/ScoredChunkResponse.md)
 - [SearchSortOrder](docs/SearchSortOrder.md)
+- [SearchTablesRequest](docs/SearchTablesRequest.md)
+- [SearchTablesResponse](docs/SearchTablesResponse.md)
 - [SearchType](docs/SearchType.md)
 - [SearchablePartType](docs/SearchablePartType.md)
 - [SeatQuotaStatus](docs/SeatQuotaStatus.md)
@@ -519,6 +527,7 @@ All URIs are relative to *http://localhost:8000*
 - [SupportedIdP](docs/SupportedIdP.md)
 - [SupportedLanguage](docs/SupportedLanguage.md)
 - [TableCellChange](docs/TableCellChange.md)
+- [TableSearchResult](docs/TableSearchResult.md)
 - [TagOrder](docs/TagOrder.md)
 - [TagResponse](docs/TagResponse.md)
 - [TemporalWorkflowStatusResponse](docs/TemporalWorkflowStatusResponse.md)
@@ -532,6 +541,8 @@ All URIs are relative to *http://localhost:8000*
 - [TenantUserOrder](docs/TenantUserOrder.md)
 - [TenantUserResponse](docs/TenantUserResponse.md)
 - [TenantUserRole](docs/TenantUserRole.md)
+- [TestConnectionRequest](docs/TestConnectionRequest.md)
+- [TestConnectionResponse](docs/TestConnectionResponse.md)
 - [TextDiff](docs/TextDiff.md)
 - [TextPart](docs/TextPart.md)
 - [TextPartOrReasoningPartOrToolPartOrDocEditPart](docs/TextPartOrReasoningPartOrToolPartOrDocEditPart.md)
@@ -540,6 +551,7 @@ All URIs are relative to *http://localhost:8000*
 - [ThreadMessageResponse](docs/ThreadMessageResponse.md)
 - [ThreadOrder](docs/ThreadOrder.md)
 - [ThreadResponse](docs/ThreadResponse.md)
+- [ToolDisplayType](docs/ToolDisplayType.md)
 - [ToolPart](docs/ToolPart.md)
 - [ToolStatus](docs/ToolStatus.md)
 - [TransferOwnerRequest](docs/TransferOwnerRequest.md)
@@ -617,7 +629,7 @@ and is automatically generated by the
 [OpenAPI Generator](https://openapi-generator.tech) project:
 
 - API version: `0.1.0`
-- Package version: `1.131.5`
+- Package version: `1.133.0`
 - Generator version: `7.21.0`
 - Build package: `org.openapitools.codegen.languages.TypeScriptFetchClientCodegen`
 

package/dist/apis/AgentApi.d.ts

@@ -1,6 +1,6 @@
 /**
  * Knowledge Stack API
- * Knowledge Stack backend API for authentication and knowledge management
+ * Knowledge Stack backend API for authentication and knowledge management.  ## Integrating (RPA / machine clients)  **Base URL.** Knowledge Stack is self-hosted — point at your own deployment host (see `servers`). The `localhost` entry is for local development only.  **Authentication.** Send `Authorization: Bearer <api-key>` on every request. Mint an API key once via `POST /v1/api-keys` from a signed-in browser session; the raw `sk-user-...` secret is returned **only** at creation, so store it then. A key inherits its owning user\'s live tenant role and path permissions — create RPA keys from a least-privilege user, and set `expires_at` for rotation. The `ks_uat` cookie scheme is browser-only and cannot be used by headless clients.  **Async work is polled, not pushed.** There are no outbound webhooks. - `POST /v1/documents/ingest` returns `201` immediately with a `workflow_id`;   poll `GET /v1/system-jobs/document_versions/{workflow_id}` until `status` is   terminal (anything other than `pending`/`processing`). The `Location` response   header points at this poll resource. - `POST /v1/workflow-runs/{run_id}/start` returns `202`; poll   `GET /v1/workflow-runs/{run_id}` until `execution_state` is `COMPLETED` or   `FAILED`. The `Location` header points at the run resource. - `POST /v1/agent/ask` is **synchronous** — it blocks until the agent finishes   and returns the answer inline. Use a generous HTTP timeout.  **Pagination.** List endpoints accept `limit`/`offset` and return `{items, total, limit, offset}`.  **Errors.** Every non-2xx body is `{detail, code, request_id}`. `code` is a stable value from a closed set (see the `ErrorResponse` schema\'s `code` enum) — branch on it rather than parsing `detail`. Quota rejections return `429` with a `Retry-After` header; transient lock contention returns a retryable `503`. Quote `request_id` (also the `x-request-id` response header) to support.  **Idempotency.** `POST /v1/workflow-runs` accepts an `idempotency_key` to dedupe retried run creation. `agent/ask` charges one message *before* running and does not refund a client-cancelled call.
  *
  * The version of the OpenAPI document: 0.1.0
  *
@@ -32,7 +32,7 @@ export interface AgentApiInterface {
      */
     agentAskRequestOpts(requestParameters: AgentAskRequest): Promise<runtime.RequestOpts>;
     /**
-     * Run a one-shot text agent request to completion and return the payload.  The request blocks until the underlying Temporal workflow finishes. Clients should set a generous HTTP timeout to accommodate tool-heavy runs.  Quota: consumes one MESSAGE before ``start_workflow``. Refund semantics distinguish cause:  * **Cancellation** (client disconnect → ``asyncio.CancelledError``,   OR explicit ``DELETE /v1/workflows/{id}`` while we await   ``handle.result()`` → Temporal-wrapped ``CancelledError``)   → **NO REFUND.** The user walked away or actively cancelled;   that\'s their volition. We still best-effort cancel the   workflow so the agent stops burning LLM tokens, but the   consume stays charged. Detection uses   ``temporalio.exceptions.is_cancelled_exception`` which spans   both forms. * Any other exception (pre-enqueue ``start_workflow`` raise,   Temporal failure, workflow-internal error) → **REFUND.** Server   / our-problem failures don\'t bill the user.
+     * Run a one-shot text agent request to completion and return the payload.  The request blocks until the underlying Temporal workflow finishes. Clients should set a generous HTTP timeout to accommodate tool-heavy runs.  Quota: consumes one MESSAGE before ``start_workflow``. Refund semantics distinguish cause:  * **Cancellation** (client disconnect → ``asyncio.CancelledError``,   OR explicit ``DELETE /v1/system-jobs/{id}`` while we await   ``handle.result()`` → Temporal-wrapped ``CancelledError``)   → **NO REFUND.** The user walked away or actively cancelled;   that\'s their volition. We still best-effort cancel the   workflow so the agent stops burning LLM tokens, but the   consume stays charged. Detection uses   ``temporalio.exceptions.is_cancelled_exception`` which spans   both forms. * Any other exception (pre-enqueue ``start_workflow`` raise,   Temporal failure, workflow-internal error) → **REFUND.** Server   / our-problem failures don\'t bill the user.
      * @summary Agent Ask Handler
      * @param {AskRequest} askRequest
      * @param {*} [options] Override http request option.
@@ -41,7 +41,7 @@ export interface AgentApiInterface {
      */
     agentAskRaw(requestParameters: AgentAskRequest, initOverrides?: RequestInit | runtime.InitOverrideFunction): Promise<runtime.ApiResponse<AskResponse>>;
     /**
-     * Run a one-shot text agent request to completion and return the payload.  The request blocks until the underlying Temporal workflow finishes. Clients should set a generous HTTP timeout to accommodate tool-heavy runs.  Quota: consumes one MESSAGE before ``start_workflow``. Refund semantics distinguish cause:  * **Cancellation** (client disconnect → ``asyncio.CancelledError``,   OR explicit ``DELETE /v1/workflows/{id}`` while we await   ``handle.result()`` → Temporal-wrapped ``CancelledError``)   → **NO REFUND.** The user walked away or actively cancelled;   that\'s their volition. We still best-effort cancel the   workflow so the agent stops burning LLM tokens, but the   consume stays charged. Detection uses   ``temporalio.exceptions.is_cancelled_exception`` which spans   both forms. * Any other exception (pre-enqueue ``start_workflow`` raise,   Temporal failure, workflow-internal error) → **REFUND.** Server   / our-problem failures don\'t bill the user.
+     * Run a one-shot text agent request to completion and return the payload.  The request blocks until the underlying Temporal workflow finishes. Clients should set a generous HTTP timeout to accommodate tool-heavy runs.  Quota: consumes one MESSAGE before ``start_workflow``. Refund semantics distinguish cause:  * **Cancellation** (client disconnect → ``asyncio.CancelledError``,   OR explicit ``DELETE /v1/system-jobs/{id}`` while we await   ``handle.result()`` → Temporal-wrapped ``CancelledError``)   → **NO REFUND.** The user walked away or actively cancelled;   that\'s their volition. We still best-effort cancel the   workflow so the agent stops burning LLM tokens, but the   consume stays charged. Detection uses   ``temporalio.exceptions.is_cancelled_exception`` which spans   both forms. * Any other exception (pre-enqueue ``start_workflow`` raise,   Temporal failure, workflow-internal error) → **REFUND.** Server   / our-problem failures don\'t bill the user.
      * Agent Ask Handler
      */
     agentAsk(requestParameters: AgentAskRequest, initOverrides?: RequestInit | runtime.InitOverrideFunction): Promise<AskResponse>;
@@ -76,12 +76,12 @@ export declare class AgentApi extends runtime.BaseAPI implements AgentApiInterfa
      */
     agentAskRequestOpts(requestParameters: AgentAskRequest): Promise<runtime.RequestOpts>;
     /**
-     * Run a one-shot text agent request to completion and return the payload.  The request blocks until the underlying Temporal workflow finishes. Clients should set a generous HTTP timeout to accommodate tool-heavy runs.  Quota: consumes one MESSAGE before ``start_workflow``. Refund semantics distinguish cause:  * **Cancellation** (client disconnect → ``asyncio.CancelledError``,   OR explicit ``DELETE /v1/workflows/{id}`` while we await   ``handle.result()`` → Temporal-wrapped ``CancelledError``)   → **NO REFUND.** The user walked away or actively cancelled;   that\'s their volition. We still best-effort cancel the   workflow so the agent stops burning LLM tokens, but the   consume stays charged. Detection uses   ``temporalio.exceptions.is_cancelled_exception`` which spans   both forms. * Any other exception (pre-enqueue ``start_workflow`` raise,   Temporal failure, workflow-internal error) → **REFUND.** Server   / our-problem failures don\'t bill the user.
+     * Run a one-shot text agent request to completion and return the payload.  The request blocks until the underlying Temporal workflow finishes. Clients should set a generous HTTP timeout to accommodate tool-heavy runs.  Quota: consumes one MESSAGE before ``start_workflow``. Refund semantics distinguish cause:  * **Cancellation** (client disconnect → ``asyncio.CancelledError``,   OR explicit ``DELETE /v1/system-jobs/{id}`` while we await   ``handle.result()`` → Temporal-wrapped ``CancelledError``)   → **NO REFUND.** The user walked away or actively cancelled;   that\'s their volition. We still best-effort cancel the   workflow so the agent stops burning LLM tokens, but the   consume stays charged. Detection uses   ``temporalio.exceptions.is_cancelled_exception`` which spans   both forms. * Any other exception (pre-enqueue ``start_workflow`` raise,   Temporal failure, workflow-internal error) → **REFUND.** Server   / our-problem failures don\'t bill the user.
      * Agent Ask Handler
      */
     agentAskRaw(requestParameters: AgentAskRequest, initOverrides?: RequestInit | runtime.InitOverrideFunction): Promise<runtime.ApiResponse<AskResponse>>;
     /**
-     * Run a one-shot text agent request to completion and return the payload.  The request blocks until the underlying Temporal workflow finishes. Clients should set a generous HTTP timeout to accommodate tool-heavy runs.  Quota: consumes one MESSAGE before ``start_workflow``. Refund semantics distinguish cause:  * **Cancellation** (client disconnect → ``asyncio.CancelledError``,   OR explicit ``DELETE /v1/workflows/{id}`` while we await   ``handle.result()`` → Temporal-wrapped ``CancelledError``)   → **NO REFUND.** The user walked away or actively cancelled;   that\'s their volition. We still best-effort cancel the   workflow so the agent stops burning LLM tokens, but the   consume stays charged. Detection uses   ``temporalio.exceptions.is_cancelled_exception`` which spans   both forms. * Any other exception (pre-enqueue ``start_workflow`` raise,   Temporal failure, workflow-internal error) → **REFUND.** Server   / our-problem failures don\'t bill the user.
+     * Run a one-shot text agent request to completion and return the payload.  The request blocks until the underlying Temporal workflow finishes. Clients should set a generous HTTP timeout to accommodate tool-heavy runs.  Quota: consumes one MESSAGE before ``start_workflow``. Refund semantics distinguish cause:  * **Cancellation** (client disconnect → ``asyncio.CancelledError``,   OR explicit ``DELETE /v1/system-jobs/{id}`` while we await   ``handle.result()`` → Temporal-wrapped ``CancelledError``)   → **NO REFUND.** The user walked away or actively cancelled;   that\'s their volition. We still best-effort cancel the   workflow so the agent stops burning LLM tokens, but the   consume stays charged. Detection uses   ``temporalio.exceptions.is_cancelled_exception`` which spans   both forms. * Any other exception (pre-enqueue ``start_workflow`` raise,   Temporal failure, workflow-internal error) → **REFUND.** Server   / our-problem failures don\'t bill the user.
      * Agent Ask Handler
      */
     agentAsk(requestParameters: AgentAskRequest, initOverrides?: RequestInit | runtime.InitOverrideFunction): Promise<AskResponse>;

package/dist/apis/AgentApi.js

@@ -3,7 +3,7 @@
 /* eslint-disable */
 /**
  * Knowledge Stack API
- * Knowledge Stack backend API for authentication and knowledge management
+ * Knowledge Stack backend API for authentication and knowledge management.  ## Integrating (RPA / machine clients)  **Base URL.** Knowledge Stack is self-hosted — point at your own deployment host (see `servers`). The `localhost` entry is for local development only.  **Authentication.** Send `Authorization: Bearer <api-key>` on every request. Mint an API key once via `POST /v1/api-keys` from a signed-in browser session; the raw `sk-user-...` secret is returned **only** at creation, so store it then. A key inherits its owning user\'s live tenant role and path permissions — create RPA keys from a least-privilege user, and set `expires_at` for rotation. The `ks_uat` cookie scheme is browser-only and cannot be used by headless clients.  **Async work is polled, not pushed.** There are no outbound webhooks. - `POST /v1/documents/ingest` returns `201` immediately with a `workflow_id`;   poll `GET /v1/system-jobs/document_versions/{workflow_id}` until `status` is   terminal (anything other than `pending`/`processing`). The `Location` response   header points at this poll resource. - `POST /v1/workflow-runs/{run_id}/start` returns `202`; poll   `GET /v1/workflow-runs/{run_id}` until `execution_state` is `COMPLETED` or   `FAILED`. The `Location` header points at the run resource. - `POST /v1/agent/ask` is **synchronous** — it blocks until the agent finishes   and returns the answer inline. Use a generous HTTP timeout.  **Pagination.** List endpoints accept `limit`/`offset` and return `{items, total, limit, offset}`.  **Errors.** Every non-2xx body is `{detail, code, request_id}`. `code` is a stable value from a closed set (see the `ErrorResponse` schema\'s `code` enum) — branch on it rather than parsing `detail`. Quota rejections return `429` with a `Retry-After` header; transient lock contention returns a retryable `503`. Quote `request_id` (also the `x-request-id` response header) to support.  **Idempotency.** `POST /v1/workflow-runs` accepts an `idempotency_key` to dedupe retried run creation. `agent/ask` charges one message *before* running and does not refund a client-cancelled call.
  *
  * The version of the OpenAPI document: 0.1.0
  *
@@ -91,7 +91,7 @@ class AgentApi extends runtime.BaseAPI {
         });
     }
     /**
-     * Run a one-shot text agent request to completion and return the payload.  The request blocks until the underlying Temporal workflow finishes. Clients should set a generous HTTP timeout to accommodate tool-heavy runs.  Quota: consumes one MESSAGE before ``start_workflow``. Refund semantics distinguish cause:  * **Cancellation** (client disconnect → ``asyncio.CancelledError``,   OR explicit ``DELETE /v1/workflows/{id}`` while we await   ``handle.result()`` → Temporal-wrapped ``CancelledError``)   → **NO REFUND.** The user walked away or actively cancelled;   that\'s their volition. We still best-effort cancel the   workflow so the agent stops burning LLM tokens, but the   consume stays charged. Detection uses   ``temporalio.exceptions.is_cancelled_exception`` which spans   both forms. * Any other exception (pre-enqueue ``start_workflow`` raise,   Temporal failure, workflow-internal error) → **REFUND.** Server   / our-problem failures don\'t bill the user.
+     * Run a one-shot text agent request to completion and return the payload.  The request blocks until the underlying Temporal workflow finishes. Clients should set a generous HTTP timeout to accommodate tool-heavy runs.  Quota: consumes one MESSAGE before ``start_workflow``. Refund semantics distinguish cause:  * **Cancellation** (client disconnect → ``asyncio.CancelledError``,   OR explicit ``DELETE /v1/system-jobs/{id}`` while we await   ``handle.result()`` → Temporal-wrapped ``CancelledError``)   → **NO REFUND.** The user walked away or actively cancelled;   that\'s their volition. We still best-effort cancel the   workflow so the agent stops burning LLM tokens, but the   consume stays charged. Detection uses   ``temporalio.exceptions.is_cancelled_exception`` which spans   both forms. * Any other exception (pre-enqueue ``start_workflow`` raise,   Temporal failure, workflow-internal error) → **REFUND.** Server   / our-problem failures don\'t bill the user.
      * Agent Ask Handler
      */
     agentAskRaw(requestParameters, initOverrides) {
@@ -102,7 +102,7 @@ class AgentApi extends runtime.BaseAPI {
         });
     }
     /**
-     * Run a one-shot text agent request to completion and return the payload.  The request blocks until the underlying Temporal workflow finishes. Clients should set a generous HTTP timeout to accommodate tool-heavy runs.  Quota: consumes one MESSAGE before ``start_workflow``. Refund semantics distinguish cause:  * **Cancellation** (client disconnect → ``asyncio.CancelledError``,   OR explicit ``DELETE /v1/workflows/{id}`` while we await   ``handle.result()`` → Temporal-wrapped ``CancelledError``)   → **NO REFUND.** The user walked away or actively cancelled;   that\'s their volition. We still best-effort cancel the   workflow so the agent stops burning LLM tokens, but the   consume stays charged. Detection uses   ``temporalio.exceptions.is_cancelled_exception`` which spans   both forms. * Any other exception (pre-enqueue ``start_workflow`` raise,   Temporal failure, workflow-internal error) → **REFUND.** Server   / our-problem failures don\'t bill the user.
+     * Run a one-shot text agent request to completion and return the payload.  The request blocks until the underlying Temporal workflow finishes. Clients should set a generous HTTP timeout to accommodate tool-heavy runs.  Quota: consumes one MESSAGE before ``start_workflow``. Refund semantics distinguish cause:  * **Cancellation** (client disconnect → ``asyncio.CancelledError``,   OR explicit ``DELETE /v1/system-jobs/{id}`` while we await   ``handle.result()`` → Temporal-wrapped ``CancelledError``)   → **NO REFUND.** The user walked away or actively cancelled;   that\'s their volition. We still best-effort cancel the   workflow so the agent stops burning LLM tokens, but the   consume stays charged. Detection uses   ``temporalio.exceptions.is_cancelled_exception`` which spans   both forms. * Any other exception (pre-enqueue ``start_workflow`` raise,   Temporal failure, workflow-internal error) → **REFUND.** Server   / our-problem failures don\'t bill the user.
      * Agent Ask Handler
      */
     agentAsk(requestParameters, initOverrides) {

package/dist/apis/ApiConnectionsApi.d.ts

@@ -1,6 +1,6 @@
 /**
  * Knowledge Stack API
- * Knowledge Stack backend API for authentication and knowledge management
+ * Knowledge Stack backend API for authentication and knowledge management.  ## Integrating (RPA / machine clients)  **Base URL.** Knowledge Stack is self-hosted — point at your own deployment host (see `servers`). The `localhost` entry is for local development only.  **Authentication.** Send `Authorization: Bearer <api-key>` on every request. Mint an API key once via `POST /v1/api-keys` from a signed-in browser session; the raw `sk-user-...` secret is returned **only** at creation, so store it then. A key inherits its owning user\'s live tenant role and path permissions — create RPA keys from a least-privilege user, and set `expires_at` for rotation. The `ks_uat` cookie scheme is browser-only and cannot be used by headless clients.  **Async work is polled, not pushed.** There are no outbound webhooks. - `POST /v1/documents/ingest` returns `201` immediately with a `workflow_id`;   poll `GET /v1/system-jobs/document_versions/{workflow_id}` until `status` is   terminal (anything other than `pending`/`processing`). The `Location` response   header points at this poll resource. - `POST /v1/workflow-runs/{run_id}/start` returns `202`; poll   `GET /v1/workflow-runs/{run_id}` until `execution_state` is `COMPLETED` or   `FAILED`. The `Location` header points at the run resource. - `POST /v1/agent/ask` is **synchronous** — it blocks until the agent finishes   and returns the answer inline. Use a generous HTTP timeout.  **Pagination.** List endpoints accept `limit`/`offset` and return `{items, total, limit, offset}`.  **Errors.** Every non-2xx body is `{detail, code, request_id}`. `code` is a stable value from a closed set (see the `ErrorResponse` schema\'s `code` enum) — branch on it rather than parsing `detail`. Quota rejections return `429` with a `Retry-After` header; transient lock contention returns a retryable `503`. Quote `request_id` (also the `x-request-id` response header) to support.  **Idempotency.** `POST /v1/workflow-runs` accepts an `idempotency_key` to dedupe retried run creation. `agent/ask` charges one message *before* running and does not refund a client-cancelled call.
  *
  * The version of the OpenAPI document: 0.1.0
  *

package/dist/apis/ApiConnectionsApi.js

@@ -3,7 +3,7 @@
 /* eslint-disable */
 /**
  * Knowledge Stack API
- * Knowledge Stack backend API for authentication and knowledge management
+ * Knowledge Stack backend API for authentication and knowledge management.  ## Integrating (RPA / machine clients)  **Base URL.** Knowledge Stack is self-hosted — point at your own deployment host (see `servers`). The `localhost` entry is for local development only.  **Authentication.** Send `Authorization: Bearer <api-key>` on every request. Mint an API key once via `POST /v1/api-keys` from a signed-in browser session; the raw `sk-user-...` secret is returned **only** at creation, so store it then. A key inherits its owning user\'s live tenant role and path permissions — create RPA keys from a least-privilege user, and set `expires_at` for rotation. The `ks_uat` cookie scheme is browser-only and cannot be used by headless clients.  **Async work is polled, not pushed.** There are no outbound webhooks. - `POST /v1/documents/ingest` returns `201` immediately with a `workflow_id`;   poll `GET /v1/system-jobs/document_versions/{workflow_id}` until `status` is   terminal (anything other than `pending`/`processing`). The `Location` response   header points at this poll resource. - `POST /v1/workflow-runs/{run_id}/start` returns `202`; poll   `GET /v1/workflow-runs/{run_id}` until `execution_state` is `COMPLETED` or   `FAILED`. The `Location` header points at the run resource. - `POST /v1/agent/ask` is **synchronous** — it blocks until the agent finishes   and returns the answer inline. Use a generous HTTP timeout.  **Pagination.** List endpoints accept `limit`/`offset` and return `{items, total, limit, offset}`.  **Errors.** Every non-2xx body is `{detail, code, request_id}`. `code` is a stable value from a closed set (see the `ErrorResponse` schema\'s `code` enum) — branch on it rather than parsing `detail`. Quota rejections return `429` with a `Retry-After` header; transient lock contention returns a retryable `503`. Quote `request_id` (also the `x-request-id` response header) to support.  **Idempotency.** `POST /v1/workflow-runs` accepts an `idempotency_key` to dedupe retried run creation. `agent/ask` charges one message *before* running and does not refund a client-cancelled call.
  *
  * The version of the OpenAPI document: 0.1.0
  *

package/dist/apis/ApiKeysApi.d.ts

@@ -1,6 +1,6 @@
 /**
  * Knowledge Stack API
- * Knowledge Stack backend API for authentication and knowledge management
+ * Knowledge Stack backend API for authentication and knowledge management.  ## Integrating (RPA / machine clients)  **Base URL.** Knowledge Stack is self-hosted — point at your own deployment host (see `servers`). The `localhost` entry is for local development only.  **Authentication.** Send `Authorization: Bearer <api-key>` on every request. Mint an API key once via `POST /v1/api-keys` from a signed-in browser session; the raw `sk-user-...` secret is returned **only** at creation, so store it then. A key inherits its owning user\'s live tenant role and path permissions — create RPA keys from a least-privilege user, and set `expires_at` for rotation. The `ks_uat` cookie scheme is browser-only and cannot be used by headless clients.  **Async work is polled, not pushed.** There are no outbound webhooks. - `POST /v1/documents/ingest` returns `201` immediately with a `workflow_id`;   poll `GET /v1/system-jobs/document_versions/{workflow_id}` until `status` is   terminal (anything other than `pending`/`processing`). The `Location` response   header points at this poll resource. - `POST /v1/workflow-runs/{run_id}/start` returns `202`; poll   `GET /v1/workflow-runs/{run_id}` until `execution_state` is `COMPLETED` or   `FAILED`. The `Location` header points at the run resource. - `POST /v1/agent/ask` is **synchronous** — it blocks until the agent finishes   and returns the answer inline. Use a generous HTTP timeout.  **Pagination.** List endpoints accept `limit`/`offset` and return `{items, total, limit, offset}`.  **Errors.** Every non-2xx body is `{detail, code, request_id}`. `code` is a stable value from a closed set (see the `ErrorResponse` schema\'s `code` enum) — branch on it rather than parsing `detail`. Quota rejections return `429` with a `Retry-After` header; transient lock contention returns a retryable `503`. Quote `request_id` (also the `x-request-id` response header) to support.  **Idempotency.** `POST /v1/workflow-runs` accepts an `idempotency_key` to dedupe retried run creation. `agent/ask` charges one message *before* running and does not refund a client-cancelled call.
  *
  * The version of the OpenAPI document: 0.1.0
  *

package/dist/apis/ApiKeysApi.js

@@ -3,7 +3,7 @@
 /* eslint-disable */
 /**
  * Knowledge Stack API
- * Knowledge Stack backend API for authentication and knowledge management
+ * Knowledge Stack backend API for authentication and knowledge management.  ## Integrating (RPA / machine clients)  **Base URL.** Knowledge Stack is self-hosted — point at your own deployment host (see `servers`). The `localhost` entry is for local development only.  **Authentication.** Send `Authorization: Bearer <api-key>` on every request. Mint an API key once via `POST /v1/api-keys` from a signed-in browser session; the raw `sk-user-...` secret is returned **only** at creation, so store it then. A key inherits its owning user\'s live tenant role and path permissions — create RPA keys from a least-privilege user, and set `expires_at` for rotation. The `ks_uat` cookie scheme is browser-only and cannot be used by headless clients.  **Async work is polled, not pushed.** There are no outbound webhooks. - `POST /v1/documents/ingest` returns `201` immediately with a `workflow_id`;   poll `GET /v1/system-jobs/document_versions/{workflow_id}` until `status` is   terminal (anything other than `pending`/`processing`). The `Location` response   header points at this poll resource. - `POST /v1/workflow-runs/{run_id}/start` returns `202`; poll   `GET /v1/workflow-runs/{run_id}` until `execution_state` is `COMPLETED` or   `FAILED`. The `Location` header points at the run resource. - `POST /v1/agent/ask` is **synchronous** — it blocks until the agent finishes   and returns the answer inline. Use a generous HTTP timeout.  **Pagination.** List endpoints accept `limit`/`offset` and return `{items, total, limit, offset}`.  **Errors.** Every non-2xx body is `{detail, code, request_id}`. `code` is a stable value from a closed set (see the `ErrorResponse` schema\'s `code` enum) — branch on it rather than parsing `detail`. Quota rejections return `429` with a `Retry-After` header; transient lock contention returns a retryable `503`. Quote `request_id` (also the `x-request-id` response header) to support.  **Idempotency.** `POST /v1/workflow-runs` accepts an `idempotency_key` to dedupe retried run creation. `agent/ask` charges one message *before* running and does not refund a client-cancelled call.
  *
  * The version of the OpenAPI document: 0.1.0
  *

package/dist/apis/AuditEventsApi.d.ts

@@ -1,6 +1,6 @@
 /**
  * Knowledge Stack API
- * Knowledge Stack backend API for authentication and knowledge management
+ * Knowledge Stack backend API for authentication and knowledge management.  ## Integrating (RPA / machine clients)  **Base URL.** Knowledge Stack is self-hosted — point at your own deployment host (see `servers`). The `localhost` entry is for local development only.  **Authentication.** Send `Authorization: Bearer <api-key>` on every request. Mint an API key once via `POST /v1/api-keys` from a signed-in browser session; the raw `sk-user-...` secret is returned **only** at creation, so store it then. A key inherits its owning user\'s live tenant role and path permissions — create RPA keys from a least-privilege user, and set `expires_at` for rotation. The `ks_uat` cookie scheme is browser-only and cannot be used by headless clients.  **Async work is polled, not pushed.** There are no outbound webhooks. - `POST /v1/documents/ingest` returns `201` immediately with a `workflow_id`;   poll `GET /v1/system-jobs/document_versions/{workflow_id}` until `status` is   terminal (anything other than `pending`/`processing`). The `Location` response   header points at this poll resource. - `POST /v1/workflow-runs/{run_id}/start` returns `202`; poll   `GET /v1/workflow-runs/{run_id}` until `execution_state` is `COMPLETED` or   `FAILED`. The `Location` header points at the run resource. - `POST /v1/agent/ask` is **synchronous** — it blocks until the agent finishes   and returns the answer inline. Use a generous HTTP timeout.  **Pagination.** List endpoints accept `limit`/`offset` and return `{items, total, limit, offset}`.  **Errors.** Every non-2xx body is `{detail, code, request_id}`. `code` is a stable value from a closed set (see the `ErrorResponse` schema\'s `code` enum) — branch on it rather than parsing `detail`. Quota rejections return `429` with a `Retry-After` header; transient lock contention returns a retryable `503`. Quote `request_id` (also the `x-request-id` response header) to support.  **Idempotency.** `POST /v1/workflow-runs` accepts an `idempotency_key` to dedupe retried run creation. `agent/ask` charges one message *before* running and does not refund a client-cancelled call.
  *
  * The version of the OpenAPI document: 0.1.0
  *
@@ -41,8 +41,8 @@ export interface AuditEventsApiInterface {
      * Creates request options for exportAuditEvents without sending the request
      * @param {string} [actorUserId] Filter to one actor
      * @param {string} [kind] Filter to one event kind
-     * @param {Date} [since] Only events at or after this timestamp
-     * @param {Date} [until] Only events strictly before this timestamp
+     * @param {Date} [since] Only events at or after this timestamp (from). Omit to export the entire audit log. Must not be in the future or after &#x60;until&#x60;.
+     * @param {Date} [until] Only events strictly before this timestamp (to). Omit for no upper bound. Must not be in the future.
      * @param {string} [subjectPathPartId] Scope to one document/folder/run subject
      * @param {boolean} [recursive] Include the subject\&#39;s descendants (needs subject)
      * @throws {RequiredError}
@@ -50,12 +50,12 @@ export interface AuditEventsApiInterface {
      */
     exportAuditEventsRequestOpts(requestParameters: ExportAuditEventsRequest): Promise<runtime.RequestOpts>;
     /**
-     * Export the tenant\'s audit events as a CSV download (admin/owner only).  Same filters as ``list_audit_events`` but unpaginated — streams every matching event newest-first. Each row resolves the actor\'s name and the subject\'s name + path so an auditor can read the file directly in Excel.
+     * Export the tenant\'s audit events as a CSV download (admin/owner only).  Same filters as ``list_audit_events`` but unpaginated — streams every matching event newest-first. Each row resolves the actor\'s name and the subject\'s name + path so an auditor can read the file directly in Excel.  With no `since`/`until` the export covers the entire audit log; pass either bound to narrow to an optional from/to window.
      * @summary Export Audit Events Handler
      * @param {string} [actorUserId] Filter to one actor
      * @param {string} [kind] Filter to one event kind
-     * @param {Date} [since] Only events at or after this timestamp
-     * @param {Date} [until] Only events strictly before this timestamp
+     * @param {Date} [since] Only events at or after this timestamp (from). Omit to export the entire audit log. Must not be in the future or after &#x60;until&#x60;.
+     * @param {Date} [until] Only events strictly before this timestamp (to). Omit for no upper bound. Must not be in the future.
      * @param {string} [subjectPathPartId] Scope to one document/folder/run subject
      * @param {boolean} [recursive] Include the subject\&#39;s descendants (needs subject)
      * @param {*} [options] Override http request option.
@@ -64,7 +64,7 @@ export interface AuditEventsApiInterface {
      */
     exportAuditEventsRaw(requestParameters: ExportAuditEventsRequest, initOverrides?: RequestInit | runtime.InitOverrideFunction): Promise<runtime.ApiResponse<Blob>>;
     /**
-     * Export the tenant\'s audit events as a CSV download (admin/owner only).  Same filters as ``list_audit_events`` but unpaginated — streams every matching event newest-first. Each row resolves the actor\'s name and the subject\'s name + path so an auditor can read the file directly in Excel.
+     * Export the tenant\'s audit events as a CSV download (admin/owner only).  Same filters as ``list_audit_events`` but unpaginated — streams every matching event newest-first. Each row resolves the actor\'s name and the subject\'s name + path so an auditor can read the file directly in Excel.  With no `since`/`until` the export covers the entire audit log; pass either bound to narrow to an optional from/to window.
      * Export Audit Events Handler
      */
     exportAuditEvents(requestParameters: ExportAuditEventsRequest, initOverrides?: RequestInit | runtime.InitOverrideFunction): Promise<Blob>;
@@ -72,8 +72,8 @@ export interface AuditEventsApiInterface {
      * Creates request options for listAuditEvents without sending the request
      * @param {string} [actorUserId] Filter to one actor
      * @param {string} [kind] Filter to one event kind
-     * @param {Date} [since] Only events at or after this timestamp
-     * @param {Date} [until] Only events strictly before this timestamp
+     * @param {Date} [since] Only events at or after this timestamp (from). Omit for no lower bound. Must not be in the future or after &#x60;until&#x60;.
+     * @param {Date} [until] Only events strictly before this timestamp (to). Omit for no upper bound. Must not be in the future.
      * @param {string} [subjectPathPartId] Scope to one document/folder/run subject
      * @param {boolean} [recursive] Include the subject\&#39;s descendants (needs subject)
      * @param {SortDirection} [sortDir] Sort by timestamp (default: DESC, newest first)
@@ -84,12 +84,12 @@ export interface AuditEventsApiInterface {
      */
     listAuditEventsRequestOpts(requestParameters: ListAuditEventsRequest): Promise<runtime.RequestOpts>;
     /**
-     * List the tenant\'s audit events, newest first (admin/owner only).  Returns every event in the caller\'s own tenant — ADMIN/OWNER bypass path permissions by design. Filter by actor, kind, time window, and/or a subject subtree. Each event carries its resolved actor name.
+     * List the tenant\'s audit events, newest first (admin/owner only).  Returns every event in the caller\'s own tenant — ADMIN/OWNER bypass path permissions by design. Filter by actor, kind, an optional `since`(from)/ `until`(to) window, and/or a subject subtree. Each event carries its resolved actor name.
      * @summary List Audit Events Handler
      * @param {string} [actorUserId] Filter to one actor
      * @param {string} [kind] Filter to one event kind
-     * @param {Date} [since] Only events at or after this timestamp
-     * @param {Date} [until] Only events strictly before this timestamp
+     * @param {Date} [since] Only events at or after this timestamp (from). Omit for no lower bound. Must not be in the future or after &#x60;until&#x60;.
+     * @param {Date} [until] Only events strictly before this timestamp (to). Omit for no upper bound. Must not be in the future.
      * @param {string} [subjectPathPartId] Scope to one document/folder/run subject
      * @param {boolean} [recursive] Include the subject\&#39;s descendants (needs subject)
      * @param {SortDirection} [sortDir] Sort by timestamp (default: DESC, newest first)
@@ -101,7 +101,7 @@ export interface AuditEventsApiInterface {
      */
     listAuditEventsRaw(requestParameters: ListAuditEventsRequest, initOverrides?: RequestInit | runtime.InitOverrideFunction): Promise<runtime.ApiResponse<PaginatedResponseEventResponse>>;
     /**
-     * List the tenant\'s audit events, newest first (admin/owner only).  Returns every event in the caller\'s own tenant — ADMIN/OWNER bypass path permissions by design. Filter by actor, kind, time window, and/or a subject subtree. Each event carries its resolved actor name.
+     * List the tenant\'s audit events, newest first (admin/owner only).  Returns every event in the caller\'s own tenant — ADMIN/OWNER bypass path permissions by design. Filter by actor, kind, an optional `since`(from)/ `until`(to) window, and/or a subject subtree. Each event carries its resolved actor name.
      * List Audit Events Handler
      */
     listAuditEvents(requestParameters: ListAuditEventsRequest, initOverrides?: RequestInit | runtime.InitOverrideFunction): Promise<PaginatedResponseEventResponse>;
@@ -115,12 +115,12 @@ export declare class AuditEventsApi extends runtime.BaseAPI implements AuditEven
      */
     exportAuditEventsRequestOpts(requestParameters: ExportAuditEventsRequest): Promise<runtime.RequestOpts>;
     /**
-     * Export the tenant\'s audit events as a CSV download (admin/owner only).  Same filters as ``list_audit_events`` but unpaginated — streams every matching event newest-first. Each row resolves the actor\'s name and the subject\'s name + path so an auditor can read the file directly in Excel.
+     * Export the tenant\'s audit events as a CSV download (admin/owner only).  Same filters as ``list_audit_events`` but unpaginated — streams every matching event newest-first. Each row resolves the actor\'s name and the subject\'s name + path so an auditor can read the file directly in Excel.  With no `since`/`until` the export covers the entire audit log; pass either bound to narrow to an optional from/to window.
      * Export Audit Events Handler
      */
     exportAuditEventsRaw(requestParameters: ExportAuditEventsRequest, initOverrides?: RequestInit | runtime.InitOverrideFunction): Promise<runtime.ApiResponse<Blob>>;
     /**
-     * Export the tenant\'s audit events as a CSV download (admin/owner only).  Same filters as ``list_audit_events`` but unpaginated — streams every matching event newest-first. Each row resolves the actor\'s name and the subject\'s name + path so an auditor can read the file directly in Excel.
+     * Export the tenant\'s audit events as a CSV download (admin/owner only).  Same filters as ``list_audit_events`` but unpaginated — streams every matching event newest-first. Each row resolves the actor\'s name and the subject\'s name + path so an auditor can read the file directly in Excel.  With no `since`/`until` the export covers the entire audit log; pass either bound to narrow to an optional from/to window.
      * Export Audit Events Handler
      */
     exportAuditEvents(requestParameters?: ExportAuditEventsRequest, initOverrides?: RequestInit | runtime.InitOverrideFunction): Promise<Blob>;
@@ -129,12 +129,12 @@ export declare class AuditEventsApi extends runtime.BaseAPI implements AuditEven
      */
     listAuditEventsRequestOpts(requestParameters: ListAuditEventsRequest): Promise<runtime.RequestOpts>;
     /**
-     * List the tenant\'s audit events, newest first (admin/owner only).  Returns every event in the caller\'s own tenant — ADMIN/OWNER bypass path permissions by design. Filter by actor, kind, time window, and/or a subject subtree. Each event carries its resolved actor name.
+     * List the tenant\'s audit events, newest first (admin/owner only).  Returns every event in the caller\'s own tenant — ADMIN/OWNER bypass path permissions by design. Filter by actor, kind, an optional `since`(from)/ `until`(to) window, and/or a subject subtree. Each event carries its resolved actor name.
      * List Audit Events Handler
      */
     listAuditEventsRaw(requestParameters: ListAuditEventsRequest, initOverrides?: RequestInit | runtime.InitOverrideFunction): Promise<runtime.ApiResponse<PaginatedResponseEventResponse>>;
     /**
-     * List the tenant\'s audit events, newest first (admin/owner only).  Returns every event in the caller\'s own tenant — ADMIN/OWNER bypass path permissions by design. Filter by actor, kind, time window, and/or a subject subtree. Each event carries its resolved actor name.
+     * List the tenant\'s audit events, newest first (admin/owner only).  Returns every event in the caller\'s own tenant — ADMIN/OWNER bypass path permissions by design. Filter by actor, kind, an optional `since`(from)/ `until`(to) window, and/or a subject subtree. Each event carries its resolved actor name.
      * List Audit Events Handler
      */
     listAuditEvents(requestParameters?: ListAuditEventsRequest, initOverrides?: RequestInit | runtime.InitOverrideFunction): Promise<PaginatedResponseEventResponse>;

package/dist/apis/AuditEventsApi.js

@@ -3,7 +3,7 @@
 /* eslint-disable */
 /**
  * Knowledge Stack API
- * Knowledge Stack backend API for authentication and knowledge management
+ * Knowledge Stack backend API for authentication and knowledge management.  ## Integrating (RPA / machine clients)  **Base URL.** Knowledge Stack is self-hosted — point at your own deployment host (see `servers`). The `localhost` entry is for local development only.  **Authentication.** Send `Authorization: Bearer <api-key>` on every request. Mint an API key once via `POST /v1/api-keys` from a signed-in browser session; the raw `sk-user-...` secret is returned **only** at creation, so store it then. A key inherits its owning user\'s live tenant role and path permissions — create RPA keys from a least-privilege user, and set `expires_at` for rotation. The `ks_uat` cookie scheme is browser-only and cannot be used by headless clients.  **Async work is polled, not pushed.** There are no outbound webhooks. - `POST /v1/documents/ingest` returns `201` immediately with a `workflow_id`;   poll `GET /v1/system-jobs/document_versions/{workflow_id}` until `status` is   terminal (anything other than `pending`/`processing`). The `Location` response   header points at this poll resource. - `POST /v1/workflow-runs/{run_id}/start` returns `202`; poll   `GET /v1/workflow-runs/{run_id}` until `execution_state` is `COMPLETED` or   `FAILED`. The `Location` header points at the run resource. - `POST /v1/agent/ask` is **synchronous** — it blocks until the agent finishes   and returns the answer inline. Use a generous HTTP timeout.  **Pagination.** List endpoints accept `limit`/`offset` and return `{items, total, limit, offset}`.  **Errors.** Every non-2xx body is `{detail, code, request_id}`. `code` is a stable value from a closed set (see the `ErrorResponse` schema\'s `code` enum) — branch on it rather than parsing `detail`. Quota rejections return `429` with a `Retry-After` header; transient lock contention returns a retryable `503`. Quote `request_id` (also the `x-request-id` response header) to support.  **Idempotency.** `POST /v1/workflow-runs` accepts an `idempotency_key` to dedupe retried run creation. `agent/ask` charges one message *before* running and does not refund a client-cancelled call.
  *
  * The version of the OpenAPI document: 0.1.0
  *
@@ -104,7 +104,7 @@ class AuditEventsApi extends runtime.BaseAPI {
         });
     }
     /**
-     * Export the tenant\'s audit events as a CSV download (admin/owner only).  Same filters as ``list_audit_events`` but unpaginated — streams every matching event newest-first. Each row resolves the actor\'s name and the subject\'s name + path so an auditor can read the file directly in Excel.
+     * Export the tenant\'s audit events as a CSV download (admin/owner only).  Same filters as ``list_audit_events`` but unpaginated — streams every matching event newest-first. Each row resolves the actor\'s name and the subject\'s name + path so an auditor can read the file directly in Excel.  With no `since`/`until` the export covers the entire audit log; pass either bound to narrow to an optional from/to window.
      * Export Audit Events Handler
      */
     exportAuditEventsRaw(requestParameters, initOverrides) {
@@ -115,7 +115,7 @@ class AuditEventsApi extends runtime.BaseAPI {
         });
     }
     /**
-     * Export the tenant\'s audit events as a CSV download (admin/owner only).  Same filters as ``list_audit_events`` but unpaginated — streams every matching event newest-first. Each row resolves the actor\'s name and the subject\'s name + path so an auditor can read the file directly in Excel.
+     * Export the tenant\'s audit events as a CSV download (admin/owner only).  Same filters as ``list_audit_events`` but unpaginated — streams every matching event newest-first. Each row resolves the actor\'s name and the subject\'s name + path so an auditor can read the file directly in Excel.  With no `since`/`until` the export covers the entire audit log; pass either bound to narrow to an optional from/to window.
      * Export Audit Events Handler
      */
     exportAuditEvents() {
@@ -175,7 +175,7 @@ class AuditEventsApi extends runtime.BaseAPI {
         });
     }
     /**
-     * List the tenant\'s audit events, newest first (admin/owner only).  Returns every event in the caller\'s own tenant — ADMIN/OWNER bypass path permissions by design. Filter by actor, kind, time window, and/or a subject subtree. Each event carries its resolved actor name.
+     * List the tenant\'s audit events, newest first (admin/owner only).  Returns every event in the caller\'s own tenant — ADMIN/OWNER bypass path permissions by design. Filter by actor, kind, an optional `since`(from)/ `until`(to) window, and/or a subject subtree. Each event carries its resolved actor name.
      * List Audit Events Handler
      */
     listAuditEventsRaw(requestParameters, initOverrides) {
@@ -186,7 +186,7 @@ class AuditEventsApi extends runtime.BaseAPI {
         });
     }
     /**
-     * List the tenant\'s audit events, newest first (admin/owner only).  Returns every event in the caller\'s own tenant — ADMIN/OWNER bypass path permissions by design. Filter by actor, kind, time window, and/or a subject subtree. Each event carries its resolved actor name.
+     * List the tenant\'s audit events, newest first (admin/owner only).  Returns every event in the caller\'s own tenant — ADMIN/OWNER bypass path permissions by design. Filter by actor, kind, an optional `since`(from)/ `until`(to) window, and/or a subject subtree. Each event carries its resolved actor name.
      * List Audit Events Handler
      */
     listAuditEvents() {

package/dist/apis/AuthApi.d.ts

@@ -1,6 +1,6 @@
 /**
  * Knowledge Stack API
- * Knowledge Stack backend API for authentication and knowledge management
+ * Knowledge Stack backend API for authentication and knowledge management.  ## Integrating (RPA / machine clients)  **Base URL.** Knowledge Stack is self-hosted — point at your own deployment host (see `servers`). The `localhost` entry is for local development only.  **Authentication.** Send `Authorization: Bearer <api-key>` on every request. Mint an API key once via `POST /v1/api-keys` from a signed-in browser session; the raw `sk-user-...` secret is returned **only** at creation, so store it then. A key inherits its owning user\'s live tenant role and path permissions — create RPA keys from a least-privilege user, and set `expires_at` for rotation. The `ks_uat` cookie scheme is browser-only and cannot be used by headless clients.  **Async work is polled, not pushed.** There are no outbound webhooks. - `POST /v1/documents/ingest` returns `201` immediately with a `workflow_id`;   poll `GET /v1/system-jobs/document_versions/{workflow_id}` until `status` is   terminal (anything other than `pending`/`processing`). The `Location` response   header points at this poll resource. - `POST /v1/workflow-runs/{run_id}/start` returns `202`; poll   `GET /v1/workflow-runs/{run_id}` until `execution_state` is `COMPLETED` or   `FAILED`. The `Location` header points at the run resource. - `POST /v1/agent/ask` is **synchronous** — it blocks until the agent finishes   and returns the answer inline. Use a generous HTTP timeout.  **Pagination.** List endpoints accept `limit`/`offset` and return `{items, total, limit, offset}`.  **Errors.** Every non-2xx body is `{detail, code, request_id}`. `code` is a stable value from a closed set (see the `ErrorResponse` schema\'s `code` enum) — branch on it rather than parsing `detail`. Quota rejections return `429` with a `Retry-After` header; transient lock contention returns a retryable `503`. Quote `request_id` (also the `x-request-id` response header) to support.  **Idempotency.** `POST /v1/workflow-runs` accepts an `idempotency_key` to dedupe retried run creation. `agent/ask` charges one message *before* running and does not refund a client-cancelled call.
  *
  * The version of the OpenAPI document: 0.1.0
  *

package/dist/apis/AuthApi.js

@@ -3,7 +3,7 @@
 /* eslint-disable */
 /**
  * Knowledge Stack API
- * Knowledge Stack backend API for authentication and knowledge management
+ * Knowledge Stack backend API for authentication and knowledge management.  ## Integrating (RPA / machine clients)  **Base URL.** Knowledge Stack is self-hosted — point at your own deployment host (see `servers`). The `localhost` entry is for local development only.  **Authentication.** Send `Authorization: Bearer <api-key>` on every request. Mint an API key once via `POST /v1/api-keys` from a signed-in browser session; the raw `sk-user-...` secret is returned **only** at creation, so store it then. A key inherits its owning user\'s live tenant role and path permissions — create RPA keys from a least-privilege user, and set `expires_at` for rotation. The `ks_uat` cookie scheme is browser-only and cannot be used by headless clients.  **Async work is polled, not pushed.** There are no outbound webhooks. - `POST /v1/documents/ingest` returns `201` immediately with a `workflow_id`;   poll `GET /v1/system-jobs/document_versions/{workflow_id}` until `status` is   terminal (anything other than `pending`/`processing`). The `Location` response   header points at this poll resource. - `POST /v1/workflow-runs/{run_id}/start` returns `202`; poll   `GET /v1/workflow-runs/{run_id}` until `execution_state` is `COMPLETED` or   `FAILED`. The `Location` header points at the run resource. - `POST /v1/agent/ask` is **synchronous** — it blocks until the agent finishes   and returns the answer inline. Use a generous HTTP timeout.  **Pagination.** List endpoints accept `limit`/`offset` and return `{items, total, limit, offset}`.  **Errors.** Every non-2xx body is `{detail, code, request_id}`. `code` is a stable value from a closed set (see the `ErrorResponse` schema\'s `code` enum) — branch on it rather than parsing `detail`. Quota rejections return `429` with a `Retry-After` header; transient lock contention returns a retryable `503`. Quote `request_id` (also the `x-request-id` response header) to support.  **Idempotency.** `POST /v1/workflow-runs` accepts an `idempotency_key` to dedupe retried run creation. `agent/ask` charges one message *before* running and does not refund a client-cancelled call.
  *
  * The version of the OpenAPI document: 0.1.0
  *

package/dist/apis/BulkDownloadApi.d.ts

@@ -1,6 +1,6 @@
 /**
  * Knowledge Stack API
- * Knowledge Stack backend API for authentication and knowledge management
+ * Knowledge Stack backend API for authentication and knowledge management.  ## Integrating (RPA / machine clients)  **Base URL.** Knowledge Stack is self-hosted — point at your own deployment host (see `servers`). The `localhost` entry is for local development only.  **Authentication.** Send `Authorization: Bearer <api-key>` on every request. Mint an API key once via `POST /v1/api-keys` from a signed-in browser session; the raw `sk-user-...` secret is returned **only** at creation, so store it then. A key inherits its owning user\'s live tenant role and path permissions — create RPA keys from a least-privilege user, and set `expires_at` for rotation. The `ks_uat` cookie scheme is browser-only and cannot be used by headless clients.  **Async work is polled, not pushed.** There are no outbound webhooks. - `POST /v1/documents/ingest` returns `201` immediately with a `workflow_id`;   poll `GET /v1/system-jobs/document_versions/{workflow_id}` until `status` is   terminal (anything other than `pending`/`processing`). The `Location` response   header points at this poll resource. - `POST /v1/workflow-runs/{run_id}/start` returns `202`; poll   `GET /v1/workflow-runs/{run_id}` until `execution_state` is `COMPLETED` or   `FAILED`. The `Location` header points at the run resource. - `POST /v1/agent/ask` is **synchronous** — it blocks until the agent finishes   and returns the answer inline. Use a generous HTTP timeout.  **Pagination.** List endpoints accept `limit`/`offset` and return `{items, total, limit, offset}`.  **Errors.** Every non-2xx body is `{detail, code, request_id}`. `code` is a stable value from a closed set (see the `ErrorResponse` schema\'s `code` enum) — branch on it rather than parsing `detail`. Quota rejections return `429` with a `Retry-After` header; transient lock contention returns a retryable `503`. Quote `request_id` (also the `x-request-id` response header) to support.  **Idempotency.** `POST /v1/workflow-runs` accepts an `idempotency_key` to dedupe retried run creation. `agent/ask` charges one message *before* running and does not refund a client-cancelled call.
  *
  * The version of the OpenAPI document: 0.1.0
  *