Agent Hub Release notes¶
Version 1.2.1 - Feb 9th 2026¶
Requires DSS >= 14.4.0
New Features¶
Short term memory: Support agents short term memory via memory fragments when enabled at agents level
Bug Fix¶
Fixed missing action buttons (copy, thumbs up/down) below the assistant’s answer
Version 1.2.0 - Feb 9th 2026¶
Requires DSS >= 14.4.0
This release includes a schema upgrade. In the event of an upgrade issue, our support team can help resolve it.
New Features¶
Human-in-the-Loop Support: Conversations with agents now include validation steps for tool actions that require explicit user confirmation, giving you more control and clarity.
Remote Database Support: Added support for remote PostgreSQL databases, enabling centralized storage of conversations, Admin settings, and related data.
Reasoning Model Visibility: Improved support for reasoning models by displaying real-time reasoning “chunks” while the agent or LLM is processing a query.
- Document Content Guardrails: Administrators can now enable guardrails within “Core Settings” to enforce content policies using forbidden words or regex. Documents that violate these policies are blocked from reaching the LLM provider.
Attachments: Forbidden documents are automatically filtered out of conversations without being deleted from the system.
Quick Agents: Documents are screened during creation; those violating policies are automatically deleted. During conversations, any agents containing forbidden documents will be filtered out.
Third-Party Agent & Tool Indicators: Third-party agents and tools (for example, Snowflake and Bedrock) are now more clearly visually distinguished in the UI.
Prompt Library Integration: The prompt library now sources content directly from collections stored in the Enterprise Asset Library of the DSS instance.
- Admin UI Improvements:
Enterprise Agents tab: Redesigned the Enterprise Agents tab in Admin Settings for enhanced usability.
Real-time Presence: Added presence badges in Admin Settings to indicate when multiple administrators are viewing the same page.
Selector components: Standardized selection components in Admin Settings and added search inputs to improve user experience.
- Interface Customization: Global interface customization is now available for all web app users, including:
White labeling: Set a default primary color, homepage title, and logos.
Asset management: Choose a managed folder for storing assets.
Agents Context: Added more information about the caller in the agents context:
dkuOnBehalfOfEmailAddressanddkuOnBehalfOfDisplayName
Bug Fix¶
Agent Sharing: Resolved an issue where group restrictions were applied even when the setting was disabled. Agents can now be shared with any user in the DSS instance when restrictions are turned off.
Traces Explorer: Fixed a bug that caused data to go missing within the traces explorer.
Draft Conversations: Disabled document uploads within draft conversations to prevent system errors.
Miscellaneous¶
Security Updates: Updated various software packages to address and resolve known security vulnerabilities.
Version 1.1.6 - Feb 9th 2026¶
Bug Fix¶
Actions restored for conversation messages - Addressed an issue where action buttons (copy, thumbs up, thumbs down, see details) would disappear after renaming a conversation and navigating between different conversations.
Version 1.1.5 - Feb 9th 2026¶
Bug Fix¶
Fixed an issue affecting agent visibility filtering that could allow users without permissions to see unintended agents. Filtering has been corrected to ensure proper access control.
Version 1.1.4 - Feb 5th 2026¶
Bug Fix¶
Fixed an issue where clicking on an agent while being in an ongoing conversation would redirect to the “Start New Conversation” page, causing the current message to be lost.
Version 1.1.3 - Jan 16th 2026¶
Bug Fix¶
Fixed an issue where users could only share agents to user belonging to their dss groups, instead by default users can share to anyone in the instance and the group restriction is opt-in (through admin settings)
Security¶
Agent Hub now includes
dkuCallerTicketinformation in the agent context, enabling tools to support end user impersonation when required. This works only with DSS ≥ 14.3.2
Version 1.1.2 - Feb 9th 2026¶
Requires DSS >= 14.2.0, or DSS >= 14.3.0 if you want to enable VLM annotation option in BYO
Bug Fix¶
Fixed an issue where table migrations failed when upgrading from v1.0.4 or earlier to v1.1.1
Resolved a security vulnerability in the Langchain core dependency (CVE-2025-68664)
Version 1.1.1 - Dec 19th 2025¶
Requires DSS >= 14.2.0, or DSS >= 14.3.0 if you want to enable VLM annotation option in BYO.
New Features¶
Added display for extraction mode in conversation attachments panel, supporting both Vision and Text modes.
Added possibility to choose in the VLM model for image attachments in admin settings
Bug Fix¶
Fixed an issue where the conversation in draft mode was cleared after switching tabs.
Fixed a bug where sources were temporarily displayed incorrectly while the agent answer was streaming.
Fixed wt1 tracking issues in standalone mode & improved WT1 event counting for the number of agents
Resolved an issue where RAG LLMs were called with the wrong ID, causing failures.
Version 1.1.0 - Dec 8th 2025¶
Requires DSS >= 14.2.0, or DSS >= 14.3.0 if you want to enable VLM annotation option in BYO.
This release introduces significant improvements to document handling (“Bring Your Own Document”), a revamped Admin Panel, enhanced sharing capabilities, and numerous UX updates.
New Features¶
- Bring Your Own Document (BYO)
If enabled by the admin, Users can upload documents directly into the conversation context. (attachment button + drag & drop)
Added admin setting for extraction modes and VLM annotation support (requires DSS >= 14.3).
Improved UI for handling and displaying documents within the current conversation.
Note: this works only with LLMs, not agents.
- Allow users to disable agents
If enabled by the admin the user can choose to directly chat with the LLM (no agents).
Chat input light redesign, moved switches and information inside the Agents button dropdown
Note: If you enable Document Upload in Conversation, users will be able to chat with the LLM. If you want to completely disable this behavior, you must disable Document Upload in Conversation
- Admin Panel & Configuration
New Admin Settings Panel: Centralized and simplified settings directly within the web application and not the Edit tab anymore.
Dynamic settings update: After admin save, the updated settings are pushed to every user and instantly reloaded
Prompt Library: Added customization options for the Prompt Library in the Admin Settings.
- Agent Sharing:
Improved Sharing Mechanism: Granular sharing options allow sharing agents only with specific groups or users within your groups.
Group Restrictions: Added “Exclude group” functionality for “My Agent” sharing.
Shared Tools for My Agents: Now Agent Hub allows the selection of the tools that have been shared from other projects and made available in the current project (Requires DSS >= v14.3.0)
Export Recipe: Added functionality for Export Recipe & Custom Dataset to export the data from the SQLite store into datasets.
JsonSnippet sources: Added handling for jsonSnippet knowledge bank responses. (multi-column retrieval)
Error Management: End users will receive an error message if an agent fails to respond or if an error occurs on the server side.
Bug Fix¶
Japanese Input: Resolved input field issues for Japanese input.
Logging: Fixed empty used_agent_ids column in database
- UI/Visual:
Fixed prompt input focus loss when creating an agent.
Fixed screenshot display for non-textual documents in Sources.
Miscellaneous¶
Security: Addressed pentesting findings and fixed dependabot alerts
Version 1.0.4 - Nov 7th 2025¶
Bug Fix¶
- Message duplication in the UI after editing
Previously, when a message was edited, it appeared twice—once before the assistant’s response was received. This duplicate is now properly removed.
- Enhanced error messages
- If the backend fails to start or another error occurs, a dialog now appears stating:
“The application failed to load. Please make sure the service is running or check the logs for more details.”
If the application loads successfully but access is restricted, an Access Denied message is displayed instead.
Optional instructions not passed to the LLM: The global system prompt was not being included in the Agent Hub completion message. This issue has been corrected.
Version 1.0.3 - Oct 24th 2025¶
Miscellaneous¶
The documentation link in the plugin description now correctly points to the latest Agent Hub guide.
Version 1.0.2 - Oct 20th 2025¶
Bug Fix¶
Settings for Augmented LLM were not being applied due to a regression. This functionality has now been re-enabled.
Version 1.0.1 - Oct 17th 2025¶
Bug Fix¶
Fixed a crash that occurred when a user tried to access a project they didn’t have permission for (which previously crashed the thread retrieving agents).
Renamed the tab in the warning message when a mandatory setting regarding “User Agents” is missing.
Miscellaneous¶
Enabled “copy/open trace” functionality for all profiles.
Changed behavior so that missing mandatory settings no longer block the save action.
Version 1.0.0 - Oct 14th 2025¶
Requires DSS >= 14.2.0
New Features¶
- Enterprise Agents
Interact with DSS Agents: Seamlessly chat with any available agent within the DSS environment, based on your web app setup.
Query Examples: Define and provide query examples to guide users start effectively.
- Quick agents (permissions based)
Create, edit & share your own agents
Prompt library: A comprehensive library to quickly kickstart the creation of your agents.
Smart mode: Automatically prefill certain fields based on your agent instructions to enhance efficiency.
- Agents Library
Access a full view of available agents, including enterprise, personal, and shared agents directly within the web app.
- Charts generation
Automated & On-Demand: Effortlessly generate charts automatically or on demand, leveraging agent artifacts in their responses.
- Monitoring
Usage Insights: Monitor agent usage effectively through the Monitoring Dashboard available in the Agent Hub.
- Dataiku Stories
Craft compelling Dataiku Stories directly from your interactions with agents.
- Feedback
Users can provide feedbacks on the responses they get
- Security tokens
Agent Hub sends user information (DSS groups, login, email) as part of the context to agents, safeguarding retrieved data.
- Agents as tool & Agents in parallel
Multi-Agent Efficiency: Query multiple agents simultaneously, using agents as tools or executing parallel calls, based on your configuration.
- Sources & Activity
View sources utilized by agents and the agents activity to have a better understanding on the generated response
View & download artifacts generated by agents
- Traces
Easily copy traces or directly access the Traces Explorer for comprehensive analysis.
- Mobile responsive design
Enjoy a seamless user experience across all devices, as the platform is fully optimized for mobile responsiveness, providing flexibility and convenience on-the-go.