Release Notes - v2.6 - June 2026

 BETA AI-Powered Reporting & Insights

Gathid v2.6 is the first release where AI capabilities are fully available (disabled by default) for all users, moving from internal testing in v2.5 to general availability (should they decide to test it out).

• Create Reports Using Natural Language - Describe what you need in plain English and Gathid will generate the report definition — entities, filters, and columns — ready to save and share. No more navigating complex configuration screens to build a new report.
• Trend-Aware AI Summaries - AI-powered summaries now highlight key trends and changes specific to each report, connecting metric movements to real-world risk and access changes so you can focus on what matters.
• Deterministic Risk Presentation - AI report summaries no longer show inconsistent risk score labels or colour cues when underlying data is unchanged, giving you a stable, reliable view during repeated reviews.
• Gathid-Guided Recommendations - AI summaries now use embedded Gathid guidelines so recommendations stay in-scope and focus on actionable, in-product steps you can take immediately.
• Derived Property Precedence in Explore - Explore queries used by AI now prioritise derived properties over raw IDs, producing prompts and summaries that better reflect how you think about your data.
• Control Which Reports Feed AI - Reports can now be flagged to indicate whether they should be included in AI-generated summaries, giving administrators full control over what feeds AI insights.
• Executive Dashboard AI Widget Controls - The AI widget on executive dashboards can be enabled or disabled per dashboard, so you can choose exactly where AI summarisation adds value.
• Reliable Executive Summary Links - Fixed an issue where Executive Summary links in AI experiences could fail, ensuring you can reliably navigate from AI insights to the relevant reports.
• Corrected Explore Query Direction - The direction of relationships in AI-generated Explore queries has been corrected so results now match expectations when stepping between entities.

Administration & Configuration

A new consolidated Configuration screen centralises setup for email, AI, and audit options — making misconfigurations easier to spot and reducing the need for support tickets.

• Consolidated Configuration Screen - Administrators now have a single Configuration screen that brings together key setup areas with clearer warnings and validation, giving you quicker setup and better visibility of potential misconfigurations before they impact users.
• Email Settings in the UI - View and configure email settings (e.g. SMTP / Microsoft Graph) directly in the application, reducing reliance on support for basic configuration checks.
• AI Configuration Controls - Opt-in reports for AI summaries, per-dashboard AI widget toggle, and compliance standards selection for AI outputs are all configurable from the new administration experience.
• Expanded Audit Configuration - The audit configuration layout has been expanded and now surfaces warnings more clearly, giving administrators earlier feedback on misconfigurations that could affect audit execution or reporting.

Search & Graph

Clearer membership views, more accurate data, and improved graph readability across entity pages and Search.

• Accurate Membership Views - Entity pages now ensure that "Member of – Direct/All" and "Membership – Direct/All" exclude entities where access has not been granted, avoiding dotted-line memberships appearing as if access were active and reducing confusion in complex graphs and compliance reviews.
• Direct Membership in Search Graph - Fixed cases where direct membership data did not appear in Search graph views, so investigators see a complete picture when tracing access paths.
• Locale-Aware Date Formatting - Dates in Search now follow your configured regional settings instead of defaulting to a single format regardless of locale.
• Pulse Widget Add/Remove Behaviour - Refined how you add and remove items from Pulse widgets on Search pages — controls are clearer and accidental removal is less likely.
• DEFINE Search – Types Without Sub-Types - A new option lets you include entities that do not have a sub-type when filtering by type in DEFINE Search, making it easier to catch "base type only" entities.

Upload Portal & File Management

The Upload Portal is now more predictable for teams that rely on manual uploads, with better visibility into upload history and reduced risk of silent overwrites or "missing file" confusion.

• Upload Date and Time Visibility - Each file now displays its upload date and time, improving auditability and making troubleshooting faster.
• Reliable Folder and File Display - Fixed an issue where creating a new folder incorrectly showed a duplicate, and ensured uploaded files appear in new folders as soon as the upload completes.
• Right-Click Paste in Empty Folders - Standard copy/paste behaviour now works everywhere, including empty folders.
• Clear Replace Warnings - A clear "replace" warning now appears when moving, copying, or cutting a file over an existing file with the same name, reducing the risk of silent overwrites.
• Accurate Delete Status - Resolved an issue where deleting files or folders could show an error even when the delete succeeded. The UI now reflects the actual outcome accurately.

Audit & Compliance

Improved handling of edge cases and clearer navigation in audit and response workflows.

• Audits With Missing Column Data - Audit definitions can now better handle scenarios where some expected columns are missing from source data, making it easier to run audits across heterogeneous or partially configured systems without constant schema changes.
• Clearer Responder Links - In responder views, URLs are now exposed directly as links instead of being hidden behind images, making navigation to evidence or supporting systems more obvious.

Systems & Stacks

Stronger resilience in stack processing and clearer relationship visualisation.

• System Names With Backslashes - Fixed a defect where system names containing a backslash could cause agents to fail. Systems can now safely use these names without impacting data loads.
• Stack Processor Resilience - The Stack Processor now detects unique constraint violations, cancels the affected job, and prevents all placeholder generation from stopping — improving resilience for long-running environments where occasional data anomalies occur.
• Focused Relationship Mapping - Relationship mapping in Systems now focuses on the selected system and its related systems, making the visualisation clearer and easier to navigate in complex environments.

Platform & Performance

Under-the-hood improvements that keep Gathid secure, stable, and compatible as your environment grows.

• Core Library Upgrades - Core libraries have been updated to stay current, improve compatibility, and reduce exposure to known vulnerabilities.
• Front-End Security Hardening - Continued defence-in-depth and secure-by-default behaviour across the front end.
• Server-Side Template Safety - Additional safeguards applied to server-side templates.
• Improved Outlook Throttling Handling - Foundational changes to better handle rate-limiting and concurrency limits when sending notifications to Microsoft Outlook for clients not leveraging noreply@gathid.com. Further tuning will continue in 2.6.x.

Questions about any of these features? Reach out to your Customer Success representative or visit support.gathid.com for the full knowledge base.