@soat/sdk 0.43.0 → 0.44.0

package/dist/index.cjs

@@ -960,7 +960,14 @@ var AiProviders = class {
 	/**
 	* Create an AI provider
 	*
-	* Creates a new LLM provider configuration
+	* Creates a new LLM provider configuration.
+	*
+	* A `bedrock` or `vertex` record must carry a credential of its own — a
+	* linked `secret_id`, or an `apiKey` in `config`. Without one the provider
+	* SDK signs with the server's own credentials (the AWS default credential
+	* chain, Google Application Default Credentials), which is refused with
+	* `400 VALIDATION_FAILED` unless the deployment allows it.
+	*
 	*/
 	static createAiProvider(options) {
 		return (options.client ?? client).post({
@@ -1018,7 +1025,7 @@ var AiProviders = class {
 	* Asks the provider which models it can run, using this provider record's own credentials and configuration, and returns provider-native model ids — the same strings `default_model` and an agent's `model` carry.
 	* Which models are reachable is a property of the credential, not of the provider type: a Vertex provider sees only the publisher models its Google Cloud project and location serve, and a Bedrock provider only the foundation models enabled in its region. Reading the list is how a caller avoids pinning a model that fails at generation time.
 	* Not every provider type can answer. `azure` lists deployments an operator named rather than models, and `ollama` lists whatever was pulled onto that host, so both return `400 MODEL_LISTING_UNSUPPORTED`.
-	* Listing resolves credentials the same way generation does, so a record that can generate can list. The API-key types (`openai`, `groq`, `xai`, `gateway`, `custom`, `anthropic`, `google`) use the record's linked secret and cannot list without one. `bedrock` and `vertex` use the linked secret when there is one — IAM keys or a Bedrock API key, a Google service-account key — and otherwise fall back to the server environment (the AWS default credential chain, Google Application Default Credentials), so a record with no `secret_id` can still list.
+	* Listing resolves credentials the same way generation does, so a record that can generate can list. The API-key types (`openai`, `groq`, `xai`, `gateway`, `custom`, `anthropic`, `google`) use the record's linked secret and cannot list without one. `bedrock` and `vertex` use the linked secret — IAM keys or a Bedrock API key, a Google service-account key. A record with no `secret_id` would fall back to the server's own credentials (the AWS default credential chain, Google Application Default Credentials); it can list only on a deployment that allows a record to use them, and returns `400 AI_PROVIDER_MISCONFIGURED` otherwise.
 	* A Vertex record needs no `config.project` when its secret is a service-account key, since the key file names its own project. A Vertex record in express mode (API key) cannot list at all: the publisher-model listing rejects API keys and needs a credential that asserts a principal, so it returns `400 MODEL_LISTING_UNSUPPORTED`.
 	* The Vertex answer is the publisher catalogue the record's `config.location` region serves. The project behind the credential is billed and quota'd for the call but does not filter the result, so a listed model may still be unavailable to that project at generation time.
 	*
@@ -3244,7 +3251,9 @@ var Quotas = class {
 	/**
 	* Create a quota
 	*
-	* Creates a project-scoped quota. `requests` is valid for `scope: project`/`api_key`; `tokens` and `cost_usd` are valid for `scope: project`/`agent`/`actor`. Any other scope/metric pair is rejected with 400 (no attribution exists to enforce it). An `actor` quota caps one end user's spend, matched from the generation's session; a null `scope_ref` means one budget *per* actor rather than a pooled project total. A `cost_usd` quota may name one `meter_type` to cap; omitting it caps every priced meter. A duplicate quota (same project, scope, scope_ref, metric, window, meter_type) is rejected with 409.
+	* Creates a project-scoped quota. `requests` is valid for `scope: project`/`api_key`; `tokens` and `cost_usd` are valid for `scope: project`/`agent`/`actor`; `storage_bytes` is valid for `scope: project` only. Any other scope/metric pair is rejected with 400 (no attribution exists to enforce it). An `actor` quota caps one end user's spend, matched from the generation's session; a null `scope_ref` means one budget *per* actor rather than a pooled project total. A `cost_usd` quota may name one `meter_type` to cap; omitting it caps every priced meter. A duplicate quota (same project, scope, scope_ref, metric, window, meter_type) is rejected with 409.
+	*
+	* `storage_bytes` caps a stored total rather than a windowed one, so it takes `window: current` and every other metric refuses that value (400 either way). It is enforced at the corpus write paths — file upload and create, document create, document ingest and re-ingest, memory-entry create — with `409 QUOTA_STORAGE_EXCEEDED` and no `Retry-After`, since no window reset clears a footprint.
 	*/
 	static createQuota(options) {
 		return (options.client ?? client).post({

package/dist/index.d.cts

@@ -3113,13 +3113,13 @@ type QuotaResourceProperties = {
   /**
    * The metric being capped
    */
-  metric: 'requests' | 'tokens' | 'cost_usd';
+  metric: 'requests' | 'tokens' | 'cost_usd' | 'storage_bytes';
   /**
-   * The window over which the metric is aggregated
+   * The window over which the metric is aggregated. storage_bytes caps a stored total rather than a windowed one, so it takes current and refuses every other value; current is refused for every other metric.
    */
-  window: 'rolling_1m' | 'rolling_1h' | 'rolling_24h' | 'calendar_month';
+  window: 'rolling_1m' | 'rolling_1h' | 'rolling_24h' | 'calendar_month' | 'current';
   /**
-   * The cap. Positive integer for requests/tokens; fractional allowed for cost_usd.
+   * The cap. Positive integer for requests/tokens/storage_bytes (bytes); fractional allowed for cost_usd.
    */
   limit: number;
   /**
@@ -5045,8 +5045,11 @@ type Quota = {
    * Public id of the api key / agent / actor the quota applies to. For `api_key` and `agent` scope, NULL means all entities of that scope type in the project. For `actor` scope, NULL means one budget *per* actor rather than a pooled total across all actors.
    */
   scope_ref?: string | null;
-  metric?: 'requests' | 'tokens' | 'cost_usd';
-  window?: 'rolling_1m' | 'rolling_1h' | 'rolling_24h' | 'calendar_month';
+  metric?: 'requests' | 'tokens' | 'cost_usd' | 'storage_bytes';
+  /**
+   * The window the metric is aggregated over; `current` on storage_bytes, which caps a stored total and never resets.
+   */
+  window?: 'rolling_1m' | 'rolling_1h' | 'rolling_24h' | 'calendar_month' | 'current';
   limit?: number;
   mode?: 'enforce' | 'monitor';
   /**
@@ -5058,7 +5061,7 @@ type Quota = {
    */
   on_unpriced?: 'block' | 'allow' | null;
   /**
-   * Current fixed-window usage for the requests metric. Null for token/cost quotas (which aggregate the usage meter at check time rather than keeping a counter) and in list responses.
+   * Current fixed-window usage for the requests metric. Null for token/cost quotas (which aggregate the usage meter at check time rather than keeping a counter), null for storage_bytes (a stored total has no window and no counter — read the footprint from the storage meter), and null in list responses.
    */
   current_usage?: {
     window_key?: string;
@@ -7643,7 +7646,7 @@ type CreateAiProviderData = {
      */
     default_model: string;
     /**
-     * Secret ID containing API credentials
+     * Secret ID containing API credentials. Required for `bedrock` and `vertex` unless `config.apiKey` carries one.
      */
     secret_id?: string;
     /**
@@ -9403,6 +9406,10 @@ type CreateDocumentErrors = {
    * Forbidden
    */
   403: ErrorResponse;
+  /**
+   * The project's `storage_bytes` quota is exceeded (`QUOTA_STORAGE_EXCEEDED`). Delete stored content, or raise the quota — no window reset clears a stored total, so no `Retry-After` is sent.
+   */
+  409: ErrorResponse;
 };
 type CreateDocumentError = CreateDocumentErrors[keyof CreateDocumentErrors];
 type CreateDocumentResponses = {
@@ -9468,7 +9475,7 @@ type IngestDocumentErrors = {
    */
   403: ErrorResponse;
   /**
-   * The file already backs a Document (a file can only be ingested once). Use `POST /documents/{document_id}/ingest` to re-process the existing document, or upload a new copy of the file to ingest it separately.
+   * The file already backs a Document (a file can only be ingested once — use `POST /documents/{document_id}/ingest` to re-process the existing document, or upload a new copy of the file to ingest it separately), or the project's `storage_bytes` quota is exceeded (`QUOTA_STORAGE_EXCEEDED`; delete stored content or raise the quota — no `Retry-After` is sent, since no window reset clears a stored total).
    */
   409: ErrorResponse;
   /**
@@ -9687,6 +9694,10 @@ type ReingestDocumentErrors = {
    * Document not found
    */
   404: ErrorResponse;
+  /**
+   * The project's `storage_bytes` quota is exceeded (`QUOTA_STORAGE_EXCEEDED`). Delete stored content, or raise the quota — no window reset clears a stored total, so no `Retry-After` is sent.
+   */
+  409: ErrorResponse;
   /**
    * The file is too large to re-ingest synchronously (`?wait=true`). Retry in background mode.
    */
@@ -10196,6 +10207,10 @@ type CreateDatasetItemErrors = {
    * Dataset not found
    */
   404: unknown;
+  /**
+   * The project's `storage_bytes` quota is exceeded (`QUOTA_STORAGE_EXCEEDED`). A fixture is a corpus write like a document, so it is bounded by the same cap; delete stored content or raise the quota — no `Retry-After` is sent, since no window reset clears a stored total.
+   */
+  409: unknown;
 };
 type CreateDatasetItemResponses = {
   /**
@@ -10248,7 +10263,7 @@ type CreateDatasetItemFromGenerationErrors = {
    */
   404: unknown;
   /**
-   * The generation has not completed, or its content was never stored or has been purged
+   * The generation has not completed, its content was never stored or has been purged, or the project's `storage_bytes` quota is exceeded (`QUOTA_STORAGE_EXCEEDED`; delete stored content or raise the quota — no `Retry-After` is sent, since no window reset clears a stored total)
    */
   409: unknown;
 };
@@ -11036,6 +11051,10 @@ type CreateFileData = {
   url: '/api/v1/files';
 };
 type CreateFileErrors = {
+  /**
+   * The project's `storage_bytes` quota is exceeded (`QUOTA_STORAGE_EXCEEDED`). Delete stored content, or raise the quota — no window reset clears a stored total, so no `Retry-After` is sent.
+   */
+  409: ErrorResponse;
   /**
    * Internal server error
    */
@@ -11089,6 +11108,10 @@ type UploadFileErrors = {
    * Insufficient permissions
    */
   403: ErrorResponse;
+  /**
+   * The project's `storage_bytes` quota is exceeded (`QUOTA_STORAGE_EXCEEDED`). Delete stored content, or raise the quota — no window reset clears a stored total, so no `Retry-After` is sent.
+   */
+  409: ErrorResponse;
 };
 type UploadFileError = UploadFileErrors[keyof UploadFileErrors];
 type UploadFileResponses = {
@@ -11117,6 +11140,10 @@ type UploadFileBase64Errors = {
    * Insufficient permissions
    */
   403: ErrorResponse;
+  /**
+   * The project's `storage_bytes` quota is exceeded (`QUOTA_STORAGE_EXCEEDED`). Delete stored content, or raise the quota — no window reset clears a stored total, so no `Retry-After` is sent.
+   */
+  409: ErrorResponse;
 };
 type UploadFileBase64Error = UploadFileBase64Errors[keyof UploadFileBase64Errors];
 type UploadFileBase64Responses = {
@@ -11175,7 +11202,7 @@ type UploadFileWithTokenErrors = {
    */
   404: ErrorResponse;
   /**
-   * Upload token already used
+   * Upload token already used, or the project's `storage_bytes` quota is exceeded (`QUOTA_STORAGE_EXCEEDED`)
    */
   409: ErrorResponse;
   /**
@@ -12974,6 +13001,10 @@ type CreateMemoryEntryErrors = {
    * Memory not found
    */
   404: unknown;
+  /**
+   * The project's `storage_bytes` quota is exceeded (`QUOTA_STORAGE_EXCEEDED`). Delete stored content, or raise the quota — no window reset clears a stored total, so no `Retry-After` is sent.
+   */
+  409: unknown;
   /**
    * Internal server error
    */
@@ -14456,13 +14487,13 @@ type CreateQuotaData = {
     /**
      * The metric being capped
      */
-    metric: 'requests' | 'tokens' | 'cost_usd';
+    metric: 'requests' | 'tokens' | 'cost_usd' | 'storage_bytes';
     /**
-     * The window over which the metric is aggregated
+     * The window over which the metric is aggregated. `current` is the only accepted value for storage_bytes (a stored total is not aggregated over time) and is refused for every other metric.
      */
-    window: 'rolling_1m' | 'rolling_1h' | 'rolling_24h' | 'calendar_month';
+    window: 'rolling_1m' | 'rolling_1h' | 'rolling_24h' | 'calendar_month' | 'current';
     /**
-     * The cap. Must be a positive integer for requests/tokens; fractional values are allowed for cost_usd.
+     * The cap. Must be a positive integer for requests/tokens/storage_bytes (bytes); fractional values are allowed for cost_usd.
      */
     limit: number;
     /**
@@ -17659,7 +17690,14 @@ declare class AiProviders {
   /**
    * Create an AI provider
    *
-   * Creates a new LLM provider configuration
+   * Creates a new LLM provider configuration.
+   *
+   * A `bedrock` or `vertex` record must carry a credential of its own — a
+   * linked `secret_id`, or an `apiKey` in `config`. Without one the provider
+   * SDK signs with the server's own credentials (the AWS default credential
+   * chain, Google Application Default Credentials), which is refused with
+   * `400 VALIDATION_FAILED` unless the deployment allows it.
+   *
    */
   static createAiProvider<ThrowOnError extends boolean = false>(options: Options<CreateAiProviderData, ThrowOnError>): RequestResult<CreateAiProviderResponses, CreateAiProviderErrors, ThrowOnError>;
   /**
@@ -17689,7 +17727,7 @@ declare class AiProviders {
    * Asks the provider which models it can run, using this provider record's own credentials and configuration, and returns provider-native model ids — the same strings `default_model` and an agent's `model` carry.
    * Which models are reachable is a property of the credential, not of the provider type: a Vertex provider sees only the publisher models its Google Cloud project and location serve, and a Bedrock provider only the foundation models enabled in its region. Reading the list is how a caller avoids pinning a model that fails at generation time.
    * Not every provider type can answer. `azure` lists deployments an operator named rather than models, and `ollama` lists whatever was pulled onto that host, so both return `400 MODEL_LISTING_UNSUPPORTED`.
-   * Listing resolves credentials the same way generation does, so a record that can generate can list. The API-key types (`openai`, `groq`, `xai`, `gateway`, `custom`, `anthropic`, `google`) use the record's linked secret and cannot list without one. `bedrock` and `vertex` use the linked secret when there is one — IAM keys or a Bedrock API key, a Google service-account key — and otherwise fall back to the server environment (the AWS default credential chain, Google Application Default Credentials), so a record with no `secret_id` can still list.
+   * Listing resolves credentials the same way generation does, so a record that can generate can list. The API-key types (`openai`, `groq`, `xai`, `gateway`, `custom`, `anthropic`, `google`) use the record's linked secret and cannot list without one. `bedrock` and `vertex` use the linked secret — IAM keys or a Bedrock API key, a Google service-account key. A record with no `secret_id` would fall back to the server's own credentials (the AWS default credential chain, Google Application Default Credentials); it can list only on a deployment that allows a record to use them, and returns `400 AI_PROVIDER_MISCONFIGURED` otherwise.
    * A Vertex record needs no `config.project` when its secret is a service-account key, since the key file names its own project. A Vertex record in express mode (API key) cannot list at all: the publisher-model listing rejects API keys and needs a credential that asserts a principal, so it returns `400 MODEL_LISTING_UNSUPPORTED`.
    * The Vertex answer is the publisher catalogue the record's `config.location` region serves. The project behind the credential is billed and quota'd for the call but does not filter the result, so a listed model may still be unavailable to that project at generation time.
    *
@@ -18835,7 +18873,9 @@ declare class Quotas {
   /**
    * Create a quota
    *
-   * Creates a project-scoped quota. `requests` is valid for `scope: project`/`api_key`; `tokens` and `cost_usd` are valid for `scope: project`/`agent`/`actor`. Any other scope/metric pair is rejected with 400 (no attribution exists to enforce it). An `actor` quota caps one end user's spend, matched from the generation's session; a null `scope_ref` means one budget *per* actor rather than a pooled project total. A `cost_usd` quota may name one `meter_type` to cap; omitting it caps every priced meter. A duplicate quota (same project, scope, scope_ref, metric, window, meter_type) is rejected with 409.
+   * Creates a project-scoped quota. `requests` is valid for `scope: project`/`api_key`; `tokens` and `cost_usd` are valid for `scope: project`/`agent`/`actor`; `storage_bytes` is valid for `scope: project` only. Any other scope/metric pair is rejected with 400 (no attribution exists to enforce it). An `actor` quota caps one end user's spend, matched from the generation's session; a null `scope_ref` means one budget *per* actor rather than a pooled project total. A `cost_usd` quota may name one `meter_type` to cap; omitting it caps every priced meter. A duplicate quota (same project, scope, scope_ref, metric, window, meter_type) is rejected with 409.
+   *
+   * `storage_bytes` caps a stored total rather than a windowed one, so it takes `window: current` and every other metric refuses that value (400 either way). It is enforced at the corpus write paths — file upload and create, document create, document ingest and re-ingest, memory-entry create — with `409 QUOTA_STORAGE_EXCEEDED` and no `Retry-After`, since no window reset clears a footprint.
    */
   static createQuota<ThrowOnError extends boolean = false>(options: Options<CreateQuotaData, ThrowOnError>): RequestResult<CreateQuotaResponses, CreateQuotaErrors, ThrowOnError>;
   /**

package/dist/index.d.mts

@@ -3113,13 +3113,13 @@ type QuotaResourceProperties = {
   /**
    * The metric being capped
    */
-  metric: 'requests' | 'tokens' | 'cost_usd';
+  metric: 'requests' | 'tokens' | 'cost_usd' | 'storage_bytes';
   /**
-   * The window over which the metric is aggregated
+   * The window over which the metric is aggregated. storage_bytes caps a stored total rather than a windowed one, so it takes current and refuses every other value; current is refused for every other metric.
    */
-  window: 'rolling_1m' | 'rolling_1h' | 'rolling_24h' | 'calendar_month';
+  window: 'rolling_1m' | 'rolling_1h' | 'rolling_24h' | 'calendar_month' | 'current';
   /**
-   * The cap. Positive integer for requests/tokens; fractional allowed for cost_usd.
+   * The cap. Positive integer for requests/tokens/storage_bytes (bytes); fractional allowed for cost_usd.
    */
   limit: number;
   /**
@@ -5045,8 +5045,11 @@ type Quota = {
    * Public id of the api key / agent / actor the quota applies to. For `api_key` and `agent` scope, NULL means all entities of that scope type in the project. For `actor` scope, NULL means one budget *per* actor rather than a pooled total across all actors.
    */
   scope_ref?: string | null;
-  metric?: 'requests' | 'tokens' | 'cost_usd';
-  window?: 'rolling_1m' | 'rolling_1h' | 'rolling_24h' | 'calendar_month';
+  metric?: 'requests' | 'tokens' | 'cost_usd' | 'storage_bytes';
+  /**
+   * The window the metric is aggregated over; `current` on storage_bytes, which caps a stored total and never resets.
+   */
+  window?: 'rolling_1m' | 'rolling_1h' | 'rolling_24h' | 'calendar_month' | 'current';
   limit?: number;
   mode?: 'enforce' | 'monitor';
   /**
@@ -5058,7 +5061,7 @@ type Quota = {
    */
   on_unpriced?: 'block' | 'allow' | null;
   /**
-   * Current fixed-window usage for the requests metric. Null for token/cost quotas (which aggregate the usage meter at check time rather than keeping a counter) and in list responses.
+   * Current fixed-window usage for the requests metric. Null for token/cost quotas (which aggregate the usage meter at check time rather than keeping a counter), null for storage_bytes (a stored total has no window and no counter — read the footprint from the storage meter), and null in list responses.
    */
   current_usage?: {
     window_key?: string;
@@ -7643,7 +7646,7 @@ type CreateAiProviderData = {
      */
     default_model: string;
     /**
-     * Secret ID containing API credentials
+     * Secret ID containing API credentials. Required for `bedrock` and `vertex` unless `config.apiKey` carries one.
      */
     secret_id?: string;
     /**
@@ -9403,6 +9406,10 @@ type CreateDocumentErrors = {
    * Forbidden
    */
   403: ErrorResponse;
+  /**
+   * The project's `storage_bytes` quota is exceeded (`QUOTA_STORAGE_EXCEEDED`). Delete stored content, or raise the quota — no window reset clears a stored total, so no `Retry-After` is sent.
+   */
+  409: ErrorResponse;
 };
 type CreateDocumentError = CreateDocumentErrors[keyof CreateDocumentErrors];
 type CreateDocumentResponses = {
@@ -9468,7 +9475,7 @@ type IngestDocumentErrors = {
    */
   403: ErrorResponse;
   /**
-   * The file already backs a Document (a file can only be ingested once). Use `POST /documents/{document_id}/ingest` to re-process the existing document, or upload a new copy of the file to ingest it separately.
+   * The file already backs a Document (a file can only be ingested once — use `POST /documents/{document_id}/ingest` to re-process the existing document, or upload a new copy of the file to ingest it separately), or the project's `storage_bytes` quota is exceeded (`QUOTA_STORAGE_EXCEEDED`; delete stored content or raise the quota — no `Retry-After` is sent, since no window reset clears a stored total).
    */
   409: ErrorResponse;
   /**
@@ -9687,6 +9694,10 @@ type ReingestDocumentErrors = {
    * Document not found
    */
   404: ErrorResponse;
+  /**
+   * The project's `storage_bytes` quota is exceeded (`QUOTA_STORAGE_EXCEEDED`). Delete stored content, or raise the quota — no window reset clears a stored total, so no `Retry-After` is sent.
+   */
+  409: ErrorResponse;
   /**
    * The file is too large to re-ingest synchronously (`?wait=true`). Retry in background mode.
    */
@@ -10196,6 +10207,10 @@ type CreateDatasetItemErrors = {
    * Dataset not found
    */
   404: unknown;
+  /**
+   * The project's `storage_bytes` quota is exceeded (`QUOTA_STORAGE_EXCEEDED`). A fixture is a corpus write like a document, so it is bounded by the same cap; delete stored content or raise the quota — no `Retry-After` is sent, since no window reset clears a stored total.
+   */
+  409: unknown;
 };
 type CreateDatasetItemResponses = {
   /**
@@ -10248,7 +10263,7 @@ type CreateDatasetItemFromGenerationErrors = {
    */
   404: unknown;
   /**
-   * The generation has not completed, or its content was never stored or has been purged
+   * The generation has not completed, its content was never stored or has been purged, or the project's `storage_bytes` quota is exceeded (`QUOTA_STORAGE_EXCEEDED`; delete stored content or raise the quota — no `Retry-After` is sent, since no window reset clears a stored total)
    */
   409: unknown;
 };
@@ -11036,6 +11051,10 @@ type CreateFileData = {
   url: '/api/v1/files';
 };
 type CreateFileErrors = {
+  /**
+   * The project's `storage_bytes` quota is exceeded (`QUOTA_STORAGE_EXCEEDED`). Delete stored content, or raise the quota — no window reset clears a stored total, so no `Retry-After` is sent.
+   */
+  409: ErrorResponse;
   /**
    * Internal server error
    */
@@ -11089,6 +11108,10 @@ type UploadFileErrors = {
    * Insufficient permissions
    */
   403: ErrorResponse;
+  /**
+   * The project's `storage_bytes` quota is exceeded (`QUOTA_STORAGE_EXCEEDED`). Delete stored content, or raise the quota — no window reset clears a stored total, so no `Retry-After` is sent.
+   */
+  409: ErrorResponse;
 };
 type UploadFileError = UploadFileErrors[keyof UploadFileErrors];
 type UploadFileResponses = {
@@ -11117,6 +11140,10 @@ type UploadFileBase64Errors = {
    * Insufficient permissions
    */
   403: ErrorResponse;
+  /**
+   * The project's `storage_bytes` quota is exceeded (`QUOTA_STORAGE_EXCEEDED`). Delete stored content, or raise the quota — no window reset clears a stored total, so no `Retry-After` is sent.
+   */
+  409: ErrorResponse;
 };
 type UploadFileBase64Error = UploadFileBase64Errors[keyof UploadFileBase64Errors];
 type UploadFileBase64Responses = {
@@ -11175,7 +11202,7 @@ type UploadFileWithTokenErrors = {
    */
   404: ErrorResponse;
   /**
-   * Upload token already used
+   * Upload token already used, or the project's `storage_bytes` quota is exceeded (`QUOTA_STORAGE_EXCEEDED`)
    */
   409: ErrorResponse;
   /**
@@ -12974,6 +13001,10 @@ type CreateMemoryEntryErrors = {
    * Memory not found
    */
   404: unknown;
+  /**
+   * The project's `storage_bytes` quota is exceeded (`QUOTA_STORAGE_EXCEEDED`). Delete stored content, or raise the quota — no window reset clears a stored total, so no `Retry-After` is sent.
+   */
+  409: unknown;
   /**
    * Internal server error
    */
@@ -14456,13 +14487,13 @@ type CreateQuotaData = {
     /**
      * The metric being capped
      */
-    metric: 'requests' | 'tokens' | 'cost_usd';
+    metric: 'requests' | 'tokens' | 'cost_usd' | 'storage_bytes';
     /**
-     * The window over which the metric is aggregated
+     * The window over which the metric is aggregated. `current` is the only accepted value for storage_bytes (a stored total is not aggregated over time) and is refused for every other metric.
      */
-    window: 'rolling_1m' | 'rolling_1h' | 'rolling_24h' | 'calendar_month';
+    window: 'rolling_1m' | 'rolling_1h' | 'rolling_24h' | 'calendar_month' | 'current';
     /**
-     * The cap. Must be a positive integer for requests/tokens; fractional values are allowed for cost_usd.
+     * The cap. Must be a positive integer for requests/tokens/storage_bytes (bytes); fractional values are allowed for cost_usd.
      */
     limit: number;
     /**
@@ -17659,7 +17690,14 @@ declare class AiProviders {
   /**
    * Create an AI provider
    *
-   * Creates a new LLM provider configuration
+   * Creates a new LLM provider configuration.
+   *
+   * A `bedrock` or `vertex` record must carry a credential of its own — a
+   * linked `secret_id`, or an `apiKey` in `config`. Without one the provider
+   * SDK signs with the server's own credentials (the AWS default credential
+   * chain, Google Application Default Credentials), which is refused with
+   * `400 VALIDATION_FAILED` unless the deployment allows it.
+   *
    */
   static createAiProvider<ThrowOnError extends boolean = false>(options: Options<CreateAiProviderData, ThrowOnError>): RequestResult<CreateAiProviderResponses, CreateAiProviderErrors, ThrowOnError>;
   /**
@@ -17689,7 +17727,7 @@ declare class AiProviders {
    * Asks the provider which models it can run, using this provider record's own credentials and configuration, and returns provider-native model ids — the same strings `default_model` and an agent's `model` carry.
    * Which models are reachable is a property of the credential, not of the provider type: a Vertex provider sees only the publisher models its Google Cloud project and location serve, and a Bedrock provider only the foundation models enabled in its region. Reading the list is how a caller avoids pinning a model that fails at generation time.
    * Not every provider type can answer. `azure` lists deployments an operator named rather than models, and `ollama` lists whatever was pulled onto that host, so both return `400 MODEL_LISTING_UNSUPPORTED`.
-   * Listing resolves credentials the same way generation does, so a record that can generate can list. The API-key types (`openai`, `groq`, `xai`, `gateway`, `custom`, `anthropic`, `google`) use the record's linked secret and cannot list without one. `bedrock` and `vertex` use the linked secret when there is one — IAM keys or a Bedrock API key, a Google service-account key — and otherwise fall back to the server environment (the AWS default credential chain, Google Application Default Credentials), so a record with no `secret_id` can still list.
+   * Listing resolves credentials the same way generation does, so a record that can generate can list. The API-key types (`openai`, `groq`, `xai`, `gateway`, `custom`, `anthropic`, `google`) use the record's linked secret and cannot list without one. `bedrock` and `vertex` use the linked secret — IAM keys or a Bedrock API key, a Google service-account key. A record with no `secret_id` would fall back to the server's own credentials (the AWS default credential chain, Google Application Default Credentials); it can list only on a deployment that allows a record to use them, and returns `400 AI_PROVIDER_MISCONFIGURED` otherwise.
    * A Vertex record needs no `config.project` when its secret is a service-account key, since the key file names its own project. A Vertex record in express mode (API key) cannot list at all: the publisher-model listing rejects API keys and needs a credential that asserts a principal, so it returns `400 MODEL_LISTING_UNSUPPORTED`.
    * The Vertex answer is the publisher catalogue the record's `config.location` region serves. The project behind the credential is billed and quota'd for the call but does not filter the result, so a listed model may still be unavailable to that project at generation time.
    *
@@ -18835,7 +18873,9 @@ declare class Quotas {
   /**
    * Create a quota
    *
-   * Creates a project-scoped quota. `requests` is valid for `scope: project`/`api_key`; `tokens` and `cost_usd` are valid for `scope: project`/`agent`/`actor`. Any other scope/metric pair is rejected with 400 (no attribution exists to enforce it). An `actor` quota caps one end user's spend, matched from the generation's session; a null `scope_ref` means one budget *per* actor rather than a pooled project total. A `cost_usd` quota may name one `meter_type` to cap; omitting it caps every priced meter. A duplicate quota (same project, scope, scope_ref, metric, window, meter_type) is rejected with 409.
+   * Creates a project-scoped quota. `requests` is valid for `scope: project`/`api_key`; `tokens` and `cost_usd` are valid for `scope: project`/`agent`/`actor`; `storage_bytes` is valid for `scope: project` only. Any other scope/metric pair is rejected with 400 (no attribution exists to enforce it). An `actor` quota caps one end user's spend, matched from the generation's session; a null `scope_ref` means one budget *per* actor rather than a pooled project total. A `cost_usd` quota may name one `meter_type` to cap; omitting it caps every priced meter. A duplicate quota (same project, scope, scope_ref, metric, window, meter_type) is rejected with 409.
+   *
+   * `storage_bytes` caps a stored total rather than a windowed one, so it takes `window: current` and every other metric refuses that value (400 either way). It is enforced at the corpus write paths — file upload and create, document create, document ingest and re-ingest, memory-entry create — with `409 QUOTA_STORAGE_EXCEEDED` and no `Retry-After`, since no window reset clears a footprint.
    */
   static createQuota<ThrowOnError extends boolean = false>(options: Options<CreateQuotaData, ThrowOnError>): RequestResult<CreateQuotaResponses, CreateQuotaErrors, ThrowOnError>;
   /**

package/dist/index.mjs

@@ -959,7 +959,14 @@ var AiProviders = class {
 	/**
 	* Create an AI provider
 	*
-	* Creates a new LLM provider configuration
+	* Creates a new LLM provider configuration.
+	*
+	* A `bedrock` or `vertex` record must carry a credential of its own — a
+	* linked `secret_id`, or an `apiKey` in `config`. Without one the provider
+	* SDK signs with the server's own credentials (the AWS default credential
+	* chain, Google Application Default Credentials), which is refused with
+	* `400 VALIDATION_FAILED` unless the deployment allows it.
+	*
 	*/
 	static createAiProvider(options) {
 		return (options.client ?? client).post({
@@ -1017,7 +1024,7 @@ var AiProviders = class {
 	* Asks the provider which models it can run, using this provider record's own credentials and configuration, and returns provider-native model ids — the same strings `default_model` and an agent's `model` carry.
 	* Which models are reachable is a property of the credential, not of the provider type: a Vertex provider sees only the publisher models its Google Cloud project and location serve, and a Bedrock provider only the foundation models enabled in its region. Reading the list is how a caller avoids pinning a model that fails at generation time.
 	* Not every provider type can answer. `azure` lists deployments an operator named rather than models, and `ollama` lists whatever was pulled onto that host, so both return `400 MODEL_LISTING_UNSUPPORTED`.
-	* Listing resolves credentials the same way generation does, so a record that can generate can list. The API-key types (`openai`, `groq`, `xai`, `gateway`, `custom`, `anthropic`, `google`) use the record's linked secret and cannot list without one. `bedrock` and `vertex` use the linked secret when there is one — IAM keys or a Bedrock API key, a Google service-account key — and otherwise fall back to the server environment (the AWS default credential chain, Google Application Default Credentials), so a record with no `secret_id` can still list.
+	* Listing resolves credentials the same way generation does, so a record that can generate can list. The API-key types (`openai`, `groq`, `xai`, `gateway`, `custom`, `anthropic`, `google`) use the record's linked secret and cannot list without one. `bedrock` and `vertex` use the linked secret — IAM keys or a Bedrock API key, a Google service-account key. A record with no `secret_id` would fall back to the server's own credentials (the AWS default credential chain, Google Application Default Credentials); it can list only on a deployment that allows a record to use them, and returns `400 AI_PROVIDER_MISCONFIGURED` otherwise.
 	* A Vertex record needs no `config.project` when its secret is a service-account key, since the key file names its own project. A Vertex record in express mode (API key) cannot list at all: the publisher-model listing rejects API keys and needs a credential that asserts a principal, so it returns `400 MODEL_LISTING_UNSUPPORTED`.
 	* The Vertex answer is the publisher catalogue the record's `config.location` region serves. The project behind the credential is billed and quota'd for the call but does not filter the result, so a listed model may still be unavailable to that project at generation time.
 	*
@@ -3243,7 +3250,9 @@ var Quotas = class {
 	/**
 	* Create a quota
 	*
-	* Creates a project-scoped quota. `requests` is valid for `scope: project`/`api_key`; `tokens` and `cost_usd` are valid for `scope: project`/`agent`/`actor`. Any other scope/metric pair is rejected with 400 (no attribution exists to enforce it). An `actor` quota caps one end user's spend, matched from the generation's session; a null `scope_ref` means one budget *per* actor rather than a pooled project total. A `cost_usd` quota may name one `meter_type` to cap; omitting it caps every priced meter. A duplicate quota (same project, scope, scope_ref, metric, window, meter_type) is rejected with 409.
+	* Creates a project-scoped quota. `requests` is valid for `scope: project`/`api_key`; `tokens` and `cost_usd` are valid for `scope: project`/`agent`/`actor`; `storage_bytes` is valid for `scope: project` only. Any other scope/metric pair is rejected with 400 (no attribution exists to enforce it). An `actor` quota caps one end user's spend, matched from the generation's session; a null `scope_ref` means one budget *per* actor rather than a pooled project total. A `cost_usd` quota may name one `meter_type` to cap; omitting it caps every priced meter. A duplicate quota (same project, scope, scope_ref, metric, window, meter_type) is rejected with 409.
+	*
+	* `storage_bytes` caps a stored total rather than a windowed one, so it takes `window: current` and every other metric refuses that value (400 either way). It is enforced at the corpus write paths — file upload and create, document create, document ingest and re-ingest, memory-entry create — with `409 QUOTA_STORAGE_EXCEEDED` and no `Retry-After`, since no window reset clears a footprint.
 	*/
 	static createQuota(options) {
 		return (options.client ?? client).post({

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@soat/sdk",
-  "version": "0.43.0",
+  "version": "0.44.0",
   "description": "TypeScript SDK for SOAT — open-source infrastructure for production-ready AI agents.",
   "keywords": [
     "ai-agents",