For the complete documentation index, see llms.txt. This page is also available as Markdown.

Changelog

Stay up to date with Emporix API and service changes.

new featureimprovement

Schema Service - cursor-based pagination for custom instances

Overview

Schema Service now supports cursor-based pagination for custom instance listing and search. Clients can use the next and prev query parameters together with the X-Next-Cursor and X-Prev-Cursor response headers to navigate large result sets without relying on deep offset pagination.

Offset pagination with pageNumber and pageSize remains supported. In cursor mode, pageNumber is ignored, X-Total-Count is not calculated or returned, and requests that send both next and prev return 400. The service also now applies deterministic sorting to all custom instance list queries by defaulting to _id:ASC when no sort is provided and appending _id:ASC as a tie-breaker when a sort is provided.

Updated endpoints

Endpoint
Description

Added support for next and prev cursor parameters and X-Next-Cursor and X-Prev-Cursor response headers for stable next and previous page navigation.

Added support for next and prev cursor parameters and X-Next-Cursor and X-Prev-Cursor response headers for stable next and previous page navigation.

Known problems

There are no known problems.

new feature

Schema Service - precision for DECIMAL attributes

Overview

Schema attributes of type DECIMAL now support an optional precision property that defines the allowed decimal step for attribute values. You can set -1 for no precision constraint, or a positive step such as 0.1, 0.01, or 0.001. When omitted for a DECIMAL attribute, the default precision 0.01 applies. For array attributes, provide precision when arrayType.type is DECIMAL.

Updated endpoints

Endpoint
Description

Creating a schema

Accepts precision on DECIMAL attributes in the request.

Updating a schema

Accepts precision on DECIMAL attributes in the request.

Retrieving a schema

Returns precision for DECIMAL attributes in the response.

Retrieving all schemas

Returns precision for DECIMAL attributes in the response.

Known problems

There are no known problems.

new feature

Import Service - operational API for import runs

Overview

The Import Service API is now available for importing external master data into Emporix. A configuration groups one or more streams; each stream extracts from a source connection, maps fields to an Emporix target type, and upserts idempotently. Imports run asynchronously and stream per-stream progress over Server-Sent Events (SSE).

This release covers the operational surface available with the importtool.import_trigger scope: reading configurations and streams, scheduling and triggering runs, monitoring and cancelling runs, and searching imported records. Administrative operations that create or change configurations, connections, streams, and mappings require the importtool.import_manage scope.

New endpoints

Endpoint
Description

Returns all import configurations defined for the tenant.

Returns a single import configuration by its identifier.

Returns the streams belonging to a configuration, ordered by sequence.

Retrieving a stream

Returns a single stream by its identifier, including resolved target types.

Retrieving a schedule

Returns the cron schedule for a configuration, or 204 when none is set.

Scheduling an import job

Creates or updates the cron schedule that runs a configuration automatically.

Triggering an import run

Starts an import run in FULL or DELTA mode, with an optional dryRun.

Retrieving run history

Returns the run history for a configuration, most recent first.

Retrieving a run

Returns a run's status together with its per-stream progress.

Streaming run progress

Streams run progress as Server-Sent Events (snapshot, stream, and final run).

Cancelling a run

Requests cancellation of an active run; force=true stops the run immediately.

Retrieving run errors

Returns the errors recorded during a run, paginated.

Returns the distinct target types that currently hold imported records.

Searches imported records of a given type with an optional search filter on the natural key.

Searches the imported records produced by a specific stream with an optional search filter.

Known problems

There are no known problems.

new feature

AI Service - ORDER support for rag_emporix tools

Overview

The rag_emporix native tool type now accepts ORDER as a predefined entityType alongside PRODUCT and custom entity types. You can configure RAG tools to search and filter against order data in the vector index.

Updated endpoints

Endpoint
Description

Listing tools

entityType of rag_emporix tools supports the predefined ORDER type.

Searching tools

entityType of rag_emporix tools supports the predefined ORDER type.

Retrieving tool by ID

entityType of rag_emporix tools supports the predefined ORDER type.

Upserting tool

entityType of rag_emporix tools supports the predefined ORDER type.

Partially updating tool

entityType of rag_emporix tools supports the predefined ORDER type.

Known problems

There are no known problems.

new feature

Indexing Service - ORDER support for reindex jobs

Overview

Reindex jobs now support ORDER as an entityType alongside PRODUCT and custom schema types. You can trigger a full reindex of order entities and track progress through the existing reindex job endpoints.

Updated endpoints

Endpoint
Description

Creating a reindex job

Accepts ORDER as a valid entityType.

Retrieving reindex jobs

Returns jobs with entityType set to ORDER.

Retrieving a reindex job

Returns jobs with entityType set to ORDER.

Known problems

There are no known problems.

improvement

Schema Service - mixin schema support for Location and Availability

Overview

Schema Service now supports LOCATION and AVAILABILITY as valid entity types for mixin schema creation and management. This allows you to define and maintain mixin schemas for location and availability entities through the Schema Service API instead of managing these schemas manually.

Updated endpoints

Endpoint
Description

Creating a schema

Added support for LOCATION and AVAILABILITY in schema types.

Updating a schema

Added support for LOCATION and AVAILABILITY in schema types.

Added support for LOCATION and AVAILABILITY in schema types.

Known problems

There are no known problems.

new feature

AI Service - managed OAuth configurations

Overview

The AI Service now provides dedicated OAuth configuration resources for self-hosted LLM authentication. You can create and manage reusable OAuth 2.0 client-credentials configurations under /agentic/oauths, then reference them from agents through llmConfig.selfHostedParams.oauth. Agent responses support expand=oauth to return the full configuration, and OAuth responses support expand=token to expand the referenced client secret token.

Currently only the client_credentials grant type is supported. Deleting an OAuth configuration that is assigned to an agent requires the force query parameter.

New endpoints

Endpoint
Description

Returns OAuth configurations for the tenant.

Searches OAuth configurations for the tenant.

Returns a single OAuth configuration by ID.

Creates or replaces an OAuth configuration.

Applies a partial update to an OAuth configuration.

Deletes an OAuth configuration. Use force=true when it is still assigned to agents.

Updated endpoints

Endpoint
Description

Retrieving agent by ID

Supports llmConfig.selfHostedParams.oauth and expand=oauth.

Listing agents

Supports llmConfig.selfHostedParams.oauth and expand=oauth.

Searching agents

Supports llmConfig.selfHostedParams.oauth and expand=oauth.

Upserting an agent

Accepts llmConfig.selfHostedParams.oauth as a reference to a managed OAuth configuration.

Accepts llmConfig.selfHostedParams.oauth as a reference to a managed OAuth configuration.

Known problems

There are no known problems.

new feature

AI Service - agent conversation listing and MS Teams tools

Overview

The AI Service API now exposes preview endpoints for listing and searching agent conversations created during MS Teams collaboration flows. Responses include conversation identifiers, context references, linked agent and session IDs, and the timestamp of the last message.

The Tools API adds the preview teams native tool type for MS Teams collaboration. Tool configuration supports teamId, tenantId, defaultInboundAgentId, and allowedOperations. Agent upsert requests can assign native tools with optional per-agent allowedOperations overrides.

New endpoints

Endpoint
Description

Listing conversations

Returns agent conversations for the tenant.

Searching conversations

Searches agent conversations for the tenant.

Updated endpoints

Endpoint
Description

Upserting tool

Supports the preview teams native tool type with config.teamId, config.tenantId, config.defaultInboundAgentId, and config.allowedOperations.

Listing tools

Responses can include teams tools with the new configuration fields.

Retrieving tool by ID

Responses can return a teams tool with the new configuration fields.

Upserting an agent

nativeTools items now accept id and optional preview allowedOperations per assigned tool.

nativeTools items now accept id and optional preview allowedOperations per assigned tool.

Known problems

There are no known problems.

new feature

Category Tree - rebuild endpoint

Overview

The Category Tree service now supports rebuilding a category tree for a given root category on demand. The rebuilt tree is returned in the response.

New endpoints

Endpoint
Description

Triggers a rebuild of the category tree for the given root category and returns the result.

Known problems

There are no known problems.

improvement

Webhook Events - restriction on order.created and order.updated

Overview

The order.created and order.updated webhook event payloads were extended with restriction.

Updated events

Event
Description

order.created

Event schema extended with restriction property.

order.updated

Event schema extended with restriction property.

Known problems

There are no known problems.

new feature

AI Rag Indexer - support for ORDER type in /filter-metadata and /rag-metadata endpoints

Overview

The endpoints /filter-metadata and /rag-metadata of AI Rag Indexer have been enhanced to support ORDER type.

Updated endpoints

Endpoint
Description

Support for ORDER type.

Support for ORDER type.

Known problems

There are no known problems.

new feature

Indexing Service - mixin path allowlist for BatteryIncluded

Overview

The IndexConfiguration object has been extended with a new optional includedMixinPaths property for the BATTERY_INCLUDED provider. It is a glob-based allowlist of product mixin paths (matched against full dot-notation paths rooted at the mixin key), where * matches within a single path segment and ** matches across one or more segments. When includedMixinPaths is non-empty, only matching mixin paths are indexed, and it must not be combined with excludedMixinKeys. Malformed glob patterns are rejected with a validation error.

Updated endpoints

Endpoint
Description

Property includedMixinPaths can be provided for the BATTERY_INCLUDED provider.

Property includedMixinPaths is returned.

Property includedMixinPaths is returned.

Property includedMixinPaths can be updated.

Known problems

There are no known problems.

new feature

Schema Service - bulk patching of custom instances

Overview

The Schema Service now supports partially updating multiple custom instances in a single request. Each item provides the custom instance id and a list of patch operations to apply. The maximum batch size is 200, and each item returns its own result status.

New endpoints

Endpoint
Description

Partially updates up to 200 custom instances in one request by applying patch operations to each item.

Known problems

There are no known problems.

new feature

Quote Service - customerReference and customerComment properties

Overview

The Quote object has been extended with two new properties that allow customers to provide additional information on their quotes:

  • customerReference - A customer-provided reference stored on the quote (e.g., a purchase order number like "PO-12345")

  • customerComment - A customer-provided comment stored on the quote (e.g., "Please deliver before Friday")

These properties can be set when creating a quote, creating a quote from a cart, or when updating a quote.

Updated endpoints

Endpoint
Description

Creating a quote

Properties customerReference and customerComment can be provided.

Properties customerReference and customerComment can be provided.

Retrieving quotes

Properties customerReference and customerComment are returned.

Retrieving a single quote

Properties customerReference and customerComment are returned.

Properties customerReference and customerComment can be updated. Customers can update these on own quotes.

Known problems

There are no known problems.

new feature

AI Service - streaming chat responses

Overview

AI Service now supports streaming LLM responses for endpoint-triggered agents through a new Server-Sent Events endpoint. This lets clients receive incremental output over text/event-stream.

New endpoints

Endpoint
Description

Adds POST /ai-service/{tenant}/agentic/chat-stream for streaming chat responses over text/event-stream.

Known problems

There are no known problems.

improvement

Schema and IAM Services - digits now supported in custom entity and scope IDs

Overview

Custom entity IDs in the Schema Service and scope IDs in the IAM Service now accept digits alongside letters and other allowed characters, giving more flexibility when naming custom types and scopes.

Updated endpoints

Endpoint
Description

The id field now accepts digits.

The id field now accepts digits.

Upserting a custom scope

The scopeId path parameter now accepts digits.

Known problems

There are no known problems.

improvement

AI Service - introduction of 'outputFormat' field

Overview

The AI Service API now accepts a JSON Schema (as a JSON string) in the outputFormat field to constrain the structure of the agent response. This enables clients to request predictable, machine-readable output for easier integration with downstream workflows.

Updated endpoints

Endpoint
Description

Retrieving agent by ID

Response now includes the outputFormat field.

Listing agents

Response now includes the outputFormat field.

Searching agents

Response now includes the outputFormat field.

Upserting an agent

Request now includes the outputFormat field.

Request now includes the outputFormat field.

Response now includes the outputFormat field.

Searching agent templates

Response now includes the outputFormat field.

improvement

Country Service - X-Version header no longer required

Overview

The X-Version header is no longer required for Country Service endpoints.

Updated endpoints

Endpoint
Description

Retrieving all countries

The X-Version header is no longer required.

Retrieving a country

The X-Version header is no longer required.

Updating a country

The X-Version header is no longer required.

Get all regions

The X-Version header is no longer required.

Retrieving a region

The X-Version header is no longer required.

Known problems

There are no known problems.

improvement

Media, Schema, and Product Services - mixin support for media assets

Overview

Media assets now support mixins, consistent with other Emporix entities. The Schema Service supports a new MEDIA schema type for defining mixin schemas, but custom attribute values must still be provided in the asset payload through mixins and metadata.mixins when creating or updating assets through the Media Service. Category and Product Service endpoints do not accept media in requests; when media is created and linked to a category or product, the media items in responses include the corresponding mixin values.

Updated endpoints

Endpoint
Description

Creating an asset

Request accepts mixins and metadata.mixins.

Updating an asset

Request accepts mixins and metadata.mixins.

Retrieving all assets

Response includes mixins and metadata.mixins.

Retrieving an asset

Response includes mixins and metadata.mixins.

Creating a schema

MEDIA added to supported schema types.

Updating a schema

MEDIA added to supported schema types.

Response media items include a mixins field.

Response media items include a mixins field.

Retrieving a product

Response media items include a mixins field.

Retrieving all products

Response media items include a mixins field.

Searching for products

Response media items include a mixins field.

Known problems

There are no known problems.

deprecated

AI Rag Indexer Service - deprecated endpoint

Overview

The AI Rag Indexer Service endpoint for reindexing has been marked as deprecated and will be removed on 2026-12-01. Instead, use the Creating reindex job endpoint.

Deprecated endpoints

Endpoint
Description

Use the Creating reindex job endpoint instead.

Known problems

There are no known problems.

deprecated

Indexing Service - deprecated endpoint

Overview

The Indexing Service endpoint for reindexing has been marked as deprecated and will be removed on 2026-12-01. Use the Creating reindex job endpoint instead.

Deprecated endpoints

Endpoint
Description

Reindexing

Use the Creating reindex job endpoint instead.

Known problems

There are no known problems.

new featureimprovement

Indexing Service - Reindex Jobs

Overview

The Indexing Service was extended with new endpoints for creating and fetching reindex jobs. For now, reindexing supports PRODUCT and custom schema entity types.

New endpoints

Endpoint
Description

Creating reindex job

Endpoint for creating reindex job.

Fetching reindex jobs

Endpoint for fetching reindex jobs.

Fetching reindex job

Endpoint for fetching reindex job.

improvement

AI Service - commerce event trigger filter support and job status update

Overview

The AI Service API now supports payload-level filtering for commerce_events agent triggers. You can define filter expressions so an agent is triggered only when an incoming commerce event payload matches the configured criteria. Additionally, jobs now support the skipped status value.

Updated types

Type
Description

AgentTrigger

Added optional filter field for the commerce_events trigger variant.

JobStatus

Added skipped enum value.

Updated endpoints

Endpoint
Description

Retrieving agent by ID

Response can now include trigger.filter for commerce_events triggers.

Listing Agents

Response can now include trigger.filter for commerce_events triggers.

Upserting agent

Request payload now supports trigger.filter for commerce_events triggers.

Partially updating agent

Request payload now supports trigger.filter for commerce_events triggers.

Listing available jobs

Job responses can now return status: skipped.

Searching jobs

Job responses can now return status: skipped.

Retrieving available job

Job response can now return status: skipped.

improvement

Category Service - X-Version header no longer required

Overview

The Category Service no longer requires the X-Version header to be passed in requests. Existing integrations that still send the X-Version header will continue to work without any changes.

Affected endpoints

All Category Service endpoints are affected. The X-Version request header is no longer required and can be safely omitted.

Endpoint
Description

Category Service API

The X-Version request header is no longer required and is ignored when provided.

Known problems

There are no known problems.

improvement

Webhook Events - cartId and quoteId on order.created and order.updated

Overview

The order.created and order.updated webhook event payloads were extended with cartId and quoteId. These fields link an order to its source cart or quote and are populated when the order was created from checkout.

  • cartId — identifier of the cart used to create the order (cart checkout).

  • quoteId — identifier of the quote from which the order was created (quote checkout).

Either field may be omitted when it does not apply to the order.

Updated events

Event
Description

order.created

Event schema extended with optional cartId and quoteId properties.

order.updated

Event schema extended with optional cartId and quoteId properties.

Known problems

There are no known problems.

major change

Overview

The Approval Service now scopes approvals to the B2B legal entity from the customer token. Previously, customers could see approvals created for other companies because the Legal-Entity-Id has not been stored on the Approval model.

New and updated approvals store legalEntity on the document and are filtered by legalEntity.id. GET responses also expose createdResource when a downstream resource (for checkout, typically an order) exists after the approval flow completes.

Behavior change (not backward compatible)

When the Legal-Entity-Id request header is sent (injected for B2B customer tokens — see B2B token and legal entity), the service:

  • Creates approvals for that legal entity only (requestor and approver contact assignments are resolved for the given entity).

  • Returns only approvals whose stored legalEntity.id matches the legal-entity-id from token.

Existing approvals created before this release do not have legalEntity stored. They are not returned when Legal-Entity-Id is present in token, because they cannot match the legal-entity filter. This is intentional and not backward compatible: tenants will not see historical approvals through the scoped B2B API.

Updated endpoints

Endpoint
Description

Creating an approval

Respects Legal-Entity-Id; persists legalEntity on the approval.

Filtered by Legal-Entity-Id when present; response includes legalEntity and optional createdResource.

Retrieving an approval

Filtered by Legal-Entity-Id when present; response includes legalEntity and optional createdResource.

Updating an approval

Filtered by Legal-Entity-Id when present.

Deleting an approval

Filtered by Legal-Entity-Id when present.

Respects Legal-Entity-Id when checking existing approvals.

Respects Legal-Entity-Id for contact assignment and approver resolution.

Schema updates

The getApprovalResponse schema was extended with:

  • legalEntity — object with id (legal entity identifier); set when the approval is created

  • createdResource — object with id (identifier of the created resource; for checkout, typically an order ID); omitted until the approved action produces a linked resource

Known problems

There are no known problems.

new featureimprovement

Emporix SDK - Batch cart item update and removal of X-Version header from category clients

Overview

A new version 1.3.1 of the Emporix SDK has been released with the following changes:

  • support for batch updating cart items via PUT /cart/{tenant}/carts/{cartId}/itemsBatch

  • removal of the X-Version header from Category Service clients (CategoryClient, CategoryAssignmentClient, CategoryTreeClient)

New methods

Method
Description

CartItemsClient.updateMultipleItemsToCart

Updates multiple cart items in a single batch operation (maximum 50 items). Returns HTTP 207 (Multi-Status) with per-item outcomes aligned to request order.

New types

Type
Description

UpdateCartItemsBatchRequest

Batch cart item update request. Each entry must include a cart item id.

UpdateCartItemsBatchEntryResponse

Per-item outcome for a batch update (index, id, code, status, message, details).

Updated types

Type
Description

BatchItemResponse

Added index field matching the position of the item in the request body.

Updated clients

Client
Description

CategoryClient

No longer sends the X-Version header on requests.

CategoryAssignmentClient

No longer sends the X-Version header on requests.

CategoryTreeClient

No longer sends the X-Version header on requests.

Known problems

There are no known problems.

new feature

AI Service - Support for file upload

Overview

The AI Service has been enhanced with a new endpoint that allows you to upload a file and attach it to the agent. Then, when calling chat endpoints it is possible to reference this file.

New endpoints

Endpoint
Description

Uploading attachment

Attaches a file to the agent

Updated endpoints

Endpoint
Description

Starting agent chat

Request body extended with attachments field.

Starting agent async chat

Request body extended with attachments field.

Known problems

There are no known problems.

new feature

AI Service - LLM models from supported providers

Overview

The AI Service has been enhanced with a new endpoint that lists available LLM models from supported LLM providers: OpenAI, Anthropic, and Google.

The service retrieves models directly from each provider's models API.

Each model entry includes an identifier, display name, optional description, and a thinking flag indicating whether the model supports thinking capabilities.

New endpoints

Endpoint
Description

Returns models grouped by provider (openai, anthropic, google). Requires ai.agent_read.

Known problems

There are no known problems.

new feature

Indexing Service - Battery Included provider support

Overview

The Indexing Service now supports Battery Included as a new index provider alongside Algolia. Only one provider can be active per tenant at a time. Battery Included only supports the MERGE site-aware fields strategy; tenants configured with the SPLIT strategy will have Battery Included indexing silently skipped.

This feature is in a preview state and payload sent to battery included is subject to change.

Updated endpoints

Endpoint
Description

Added BATTERY_INCLUDED provider support with new excludedMixinKeys field. Only one provider can be active per tenant at a time.

Response may include Battery Included configurations with excludedMixinKeys.

provider path parameter now accepts BATTERY_INCLUDED.

Updating a configuration

Supports updating Battery Included configurations with excludedMixinKeys.

Response may include Battery Included public configurations.

provider path parameter now accepts BATTERY_INCLUDED.

Known problems

There are no known problems.

deprecated

Availability Service - deprecated endpoints

Overview

The Availability Service endpoints for managing availabilities that accept site as a query parameter, which were marked as deprecated on 2025-01-13, will be removed on 2026-09-01. They have been replaced with new endpoints that accept site as a path parameter. The Availability Service endpoints for managing locations have been marked as deprecated and will be removed on 2026-09-01 due to the planned sunset of the related functionalities.

Deprecated endpoints

Availability endpoints accepting site as query parameter

Endpoint
Description

[GET] /{tenant}/availability/{productId}?site

Deprecated. Use Retrieving a product availability instead.

[POST] /{tenant}/availability/{productId}?site

Deprecated. Use Creating a new availability for a product instead.

[PUT] /{tenant}/availability/{productId}?site

Deprecated. Use Upserting availability information of a product instead.

[DELETE] /{tenant}/availability/{productId}?site

Deprecated. Use Deleting availability information of a product instead.

[GET] /{tenant}/availability?site

Deprecated. Use Retrieving all availability information for a site instead.

[POST] /{tenant}/availability?site

Deprecated. Use Retrieving product availabilities instead.

Deprecated due to planned sunset of related functionalities.

Deprecated due to planned sunset of related functionalities.

Updating locations

Deprecated due to planned sunset of related functionalities.

Deleting a location

Deprecated due to planned sunset of related functionalities.

Deprecated due to planned sunset of related functionalities.

Known problems

There are no known problems.

deprecated

Supplier Service - deprecation

Overview

The Supplier Service has been deprecated and will be removed on 2026-09-01.

Known problems

There are no known problems.

improvement

Schema Service — explicit owner on custom instance creation

Overview

The Schema Service now accepts an optional owner field in the request body when creating a custom instance, so clients can assign ownership explicitly instead of relying on the platform to derive it from the calling principal.

The owner field can be set explicitly only by clients authorized with a manage scope (schema.custominstance_manage or custom.{lowerCaseType}_manage). When the request is authorized with a manage_own scope, the field is ignored and ownership is set automatically to the user who created the custom instance. When a manage scope is granted but the field is omitted, ownership is likewise set automatically to the user who created the custom instance.

The following limitations apply when providing owner:

  • type can only be set to CUSTOMER or EMPLOYEE. The SERVICE type cannot be assigned explicitly.

  • legalEntityId can be provided only when type is CUSTOMER.

  • The owner field is immutable. It can only be set during creation.

Updated endpoints

Endpoint
Description

The request body now accepts an optional owner.

The request body now accepts an optional owner field.

Known problems

There are no known problems.

new feature

AI Service - RAG_EMPORIX support for custom entity types

Overview

The tool of RAG_EMPORIX type allow specifying the custom entity type by using entityType field.

Updated endpoints

Endpoint
Description

Listing tools

Field entityType of RAG_EMPORIX tool type supports custom entity types.

Searching tools

Field entityType of RAG_EMPORIX tool type supports custom entity types.

Retrieving tool by ID

Field entityType of RAG_EMPORIX tool type supports custom entity types.

Upserting tool

Field entityType of RAG_EMPORIX tool type supports custom entity types.

Partially updating tool

Field entityType of RAG_EMPORIX tool type supports custom entity types.

Known problems

There are no known problems.

improvement

Webhook Events - updated schema.custom-instance-updated and schema.custom-instance-created events schema

Overview

The schema.custom-instance-updated and schema.custom-instance-created webhook events documentation has been updated to reflect the current payload structure.

Updated event

Event
Description

schema.custom-instance-updated

Updated event schema by including nested media and owner data.

schema.custom-instance-created

Updated event schema by including nested media and owner data.

Known problems

There are no known problems.

improvement

Approval Service - quote approval process

Overview

The Approval Service supports quote checkout approvals alongside cart approvals. For resourceType: QUOTE, the client sends the quote ID in resourceId and does not send details. The service loads quote line and price data from the quote integration. For resourceType: CART, details remains required as before.

Updated endpoints

Endpoint
Description

Creating an approval

Supports resourceType: QUOTE with quote resourceId.

Response resource payloads align with updated item and price schemas.

Retrieving an approval

Response resource payloads align with updated item and price schemas.

Request may use resourceType: QUOTE and a quote resourceId.

Request may use resourceType: QUOTE and a quote resourceId.

Schema updates

The Approval Service API reference schemas were extended with:

  • resourceType value QUOTE

  • resourceItem: itemId; itemPrice as line-level price (unitPrice, newUnitPrice, calculatedPrice)

  • price (totals): netValue, grossValue, taxValue

Known problems

There are no known problems.

deprecated

Approval Service - deprecated fields

Overview

All deprecations will be removed on 2026-11-30. Several fields in the approval response have been deprecated in favor of new, more descriptive fields. The new fields provide a more comprehensive and consistent way to access price calculations.

Deprecated approval resource level fields:

  • totalPrice.amount - use totalPrice.netValue, totalPrice.grossValue and totalPrice.taxValue instead

  • subTotalPrice.amount - use subTotalPrice.netValue, subTotalPrice.grossValue and subTotalPrice.taxValue instead

Deprecated approval resource-item level fields:

  • itemYrn - use itemId instead

  • itemPrice.amount - use calculatedPrice and unitPrice instead

Updated endpoints

Endpoint
Description

Deprecated response fields in favor of new more comprehensive way to represent prices.

Retrieving an approval

Deprecated response fields in favor of new more comprehensive way to represent prices.

Known problems

There are no known problems.

new feature

AI Rag Indexer - support for custom entity types in /filter-metadata and /rag-metadata endpoints

Overview

The endpoints /filter-metadata and /rag-metadata of AI Rag Indexer have been enhanced to support custom entity types.

Updated endpoints

Endpoint
Description

Support for custom entity types.

Support for custom entity types.

Known problems

There are no known problems.

deprecated

AI Rag Indexer - name and description fields from /filter-metadata endpoint deprecated

Overview

The fields name and description from the response of /filter-metadata endpoint of AI Rag Indexer have been deprecated. These fields were previously used to provide additional information about the filters, but they are no longer necessary for the functionality of the service. They will be removed on 2026-12-01.

Affected endpoint

Endpoint
Description

Fields name and description from the response are marked as deprecated.

Known problems

There are no known problems.

new feature

AI Service - RAG_EMPORIX filter fields

Overview

The tool of RAG_EMPORIX type has been enhanced with a filterFields field, which allows specifying which fields are available for filtering in the vector index.

Updated endpoints

Endpoint
Description

Listing tools

Response enhanced with filterFields field.

Searching tools

Response enhanced with filterFields field.

Retrieving tool by ID

Response enhanced with filterFields field.

Upserting tool

Request body enhanced with filterFields field.

Partially updating tool

Request body enhanced with filterFields field.

Known problems

There are no known problems.

deprecated

SEPA Export Service - deprecation

Overview

The SEPA Export Service has been deprecated and will be removed on 2026-08-24.

Known problems

There are no known problems.

Overview

The SEPA Export Service is now deprecated in its entirety. All endpoints exposed by the service are deprecated and the service is scheduled to be officially removed on 2026-08-24.

Deprecated endpoints

All endpoints under the /sepa-export/{tenant}/ base path are deprecated. For the full list of affected endpoints, refer to the SEPA Export Service API Reference.

Known problems

There are no known problems.

deprecated

Pick-pack Service - service deprecation

Overview

The Pick-pack Service is now deprecated in its entirety. All endpoints exposed by the service are deprecated and the service is scheduled to be officially removed on 2026-08-24.

Deprecated endpoints

All endpoints under the /pick-pack/{tenant} base path are deprecated, including those used to manage pick-pack orders, order cycles, assignees, recalculations, and events. For the full list of affected endpoints, refer to the Pick-pack Service API Reference.

Known problems

There are no known problems.

new feature

Overview

The Price Service search endpoint now supports a new siteCodes field, which allows searching for prices across multiple site codes in a single request. If the existing siteCode field is specified, siteCodes is ignored.

Updated endpoints

Endpoint
Description

Searching for prices

New attribute siteCodes added to request payload. Accepts an array of site codes for which the prices should be retrieved.

Known problems

There are no known problems.

new feature

Emporix SDK - Dynamic variants support and recalculation jobs

Overview

A new version 1.3.0 of the Emporix SDK adds support for DYNAMIC_VARIANT products and introduces a dedicated client for asynchronous dynamic variant recalculation jobs.

Dynamic variants form a flexible hierarchy (up to 4 levels). Each variant stores its own distinguishing attributes (ownVariantAttributes) and a root dynamic variant contains a denormalized variants map of all descendants. For bulk imports, the variant tree can be rebuilt asynchronously using the recalculation job endpoints.

New client

Client
Description

RecalculationJobClient

Declarative HTTP client for Product Service dynamic variant recalculation jobs. Automatically registered as a Spring bean.

New methods

Method
Description

RecalculationJobClient.triggerDynamicVariantRecalculation

Triggers asynchronous recalculation jobs for one or more dynamic variants.

RecalculationJobClient.listDynamicVariantRecalculationJobs

Lists recalculation jobs, optionally filtered by status.

RecalculationJobClient.getDynamicVariantRecalculationJob

Retrieves a single recalculation job by ID.

New types

Dynamic variant recalculation

Type
Description

DynamicVariantRecalculationRequest

Request body for triggering recalculation. Contains up to 1000 productIds.

DynamicVariantRecalculationResponse

Response containing jobs and skippedProductIds.

DynamicVariantRecalculationJobResponse

Represents a single recalculation job (id, root, status, timestamps, error).

DynamicVariantRecalculationJobStatus

Job status enum: PENDING, PROCESSING, FINISHED, FAILED, FAILED_PERMANENT.

Dynamic variant attributes and variant map

Type
Description

DynamicVariantAttribute

Attribute entry with localized name and typed value.

DynamicVariantAttributeValue

Typed value containing type, qualifier, optional unit, and localized display name.

DynamicVariantAttributeType

Attribute value type enum (e.g. STRING, BOOLEAN, NUMBER, DECIMAL).

VariantAttributeQualifier

Qualifier value with custom serializer/deserializer for stable JSON representation.

VariantMapEntry

Entry stored in a root product’s variants map with accumulated attributes for a descendant.

DynamicVariantInfo

Metadata information for dynamic variant trees (for example, missing ancestor, cycle detection).

Requests for DYNAMIC_VARIANT products

Type
Description

DynamicVariantCreateRequest

Create request for DYNAMIC_VARIANT products, including ownVariantAttributes.

DynamicVariantUpdateRequest

Update request for DYNAMIC_VARIANT products, including ownVariantAttributes.

Product presentation - brand & labels

Type
Description

Brand

Brand representation embedded in ProductResponse (resolved when expanded by the API).

Label

Label representation embedded in ProductResponse (resolved when expanded by the API).

Updated types

Type
Change summary

ProductResponse

Added dynamic variant support (dynamicVariantType, parentVariantId, parentVariantPath, sellable, ownVariantAttributes, inheritedVariantAttributes, root variants map) and presentation fields (labelIds, labels, brandId, brand).

ProductMetadataResponse

Added dynamicVariantInfo for dynamic-variant tree diagnostics.

GenericProductCreateRequest

Added dynamic variant fields (productType, parentVariantId, dynamicVariantType, sellable, ownVariantAttributes).

GenericProductUpdateRequest

Added dynamic variant fields (sellable, dynamicVariantType, parentVariantId, ownVariantAttributes).

ProductPartialUpdateRequest

Added dynamic variant fields (dynamicVariantType, sellable, ownVariantAttributes) and additional partial-update fields (labelIds, brandId).

Known problems

Constructor signature changes (compile-time breaking for some usages)

The all-args constructors generated by Lombok changed due to newly added fields in:

  • ProductResponse

  • GenericProductCreateRequest

  • GenericProductUpdateRequest

  • ProductPartialUpdateRequest

If your code instantiates these classes using constructors, you may hit compilation errors after upgrading.

Recommended fix: use the Lombok builder (.builder()...build()) instead of constructors.

Class GenericProductUpdateRequest is now using MetadataRequest instead of VariantProductMetadataRequest for the metadata field.

The following classes were updated to contain productType field that is necessary for upserting and creating products:

  • BundledProductCreateRequest

  • DynamicVariantCreateRequest