Kiss release
This section includes key information regarding new features included in Aura Platform for Kiss and main changes in comparison with previous releases
What’s new in Kiss (9.7.0)?
Discover here the new features and modifications at a glance. You can find the technical details in the release changelog and the Release Notes.
| Novelties | Description |
|---|---|
| Write calls to OpenAI and consumed tokens in logs | ✔️ As a first step to improve ATRIA monitoring, the use of the same application received from atria-model-gateway in RAG allows including all the calls to OpenAI and consumed tokens in logs |
| Deactivate authentication flows for Aiwi channel in ES | ✔️ Aiwi channel becomes an anonymous channel, after deactivating all its login flows |
| Aura components versions upgrade | ✔️ Aura components versions have been upgraded, improving performance and ensuring compatibility with the latest features and security standards |
| Available disambiguation score for channels and applications | ✔️ Added flag to return the score of the different disambiguation options, so that channels and applications can process it |
| Modifying ATRIA configuration | ✔️ The guidelines for the modification of ATRIA configuration and documents uploading have been completed, including all the feasible processes |
| Improved ATRIA applications documentation | ✔️ The ATRIA applications descriptive document has been updated and completed with advanced details: applications types, applications model, etc. |
| Added example cases for ATRIA capabilities | ✔️ Practical example cases for ATRIA Generative and RAG, in order to ease the understanding of these capabilities |
| More efficient ATRIA RAG Generate DB documents processing | ✔️ Added a record of the documents current state, allowing only the processing of updates or additions |
| URM Global script Kernel configuration | ✔️ New guidelines for the configuration of the URM Global script in Kernel |
| Improved Aura documentation | ✔️ New documentation blocks in Aura Technical Documentation website ✔️ Corrections in CSV entities documentation ✔️ New tags and categories and review of existing ones ✔️ Documents are continuously updated, reviewed and corrected in order to achieve, release after release, a better quality in Aura Platform documentation |
| UserID in anonymous users | ✔️ Due to Kernel requirements, now anonymous users have userId and auraGlobalId, as these values are needed for the generation of KPIs. |
| Avoid corrupted files in KPIs | ✔️ Changed default value of variables for managing KPIs blobs in different Aura components to avoid duplicate headers and, consequently, prevent corrupted files |
| Improvements in Aura monitoring | ✔️ Added aura-file-manager metrics ✔️ Included labels to identify the channel and Living App of the request in certain metrics. This will help measure traffic impact in case of issue ✔️ Aura HTTP inbound/outbound dashboards are now generated by channel |
| Aura Databricks Jobs operation | ✔️ Improved operation of the component in charge of importing Avro datasets into Kernel |
| General guidelines for Kernel configuration | ✔️ New guidelines outlining the orderly steps for Kernel configuration, applicable to all Aura features and tools |
| Improved aura-configuration-api logs handling | ✔️ Added correlator information to logs in the aura-configuration-api to improve traceability and debugging |
| Improvements in aura-file-manager | ✔️ Added aura-file-manager metrics ✔️ Redis used by default for storage ✔️ Updates in aura-file-manager documentation |
| Modifications in aura-kpis-uploader | ✔️ Add version control for Avro Adapter models ✔️ Added new environment variables for an improved management of CSV and Avro files and entities ✔️ Improved documentation, with comprehensive details of processes |
| Deactivation of Aura API request dataset uploading | ✔️ This CSV-based entity is no longer uploaded to Kernel, as it is deprecated. Its information will be registered in a new URM Avro entity |
