Skip to main content

Cogynt.ai v2.22 Release Notes

· 3 min read

Cogility is delighted to announce that Cogynt.ai version 2.22 is generally available for AWS customers as of June 2026. See what's new below.

For more information, or to have your Cogynt.ai instance updated, please contact us.

Authoring and HCEP​

Support HTTP Request in Event Patterns​

  • Users can now make event-driven HTTP requests via Event Patterns.
  • Users can adjust and customize the following HTTP request settings:
    • Request Per Minute (for Rate Limiting)
    • Max concurrent request
    • API Caching
    • Cache Priority
    • Cache TTL
    • TLS/SSL Encryption
  • Users can define HTTP query and header parameters, as well as a response schema in Event Types that reference HTTP data sources.
  • Users will now have beta access to LLM support in Authoring.
  • Dedicated API endpoint was added to Intelligence OTP for Authoring to provide model selection, custom prompts and in-line reference to mapped parameters.

Enhanced TLS/SSL Support​

  • TLS is now supported in Pinot and Postgres.
  • Expanded support for PKCS#12 cert formats for all data sources.
  • Expanded support for PEM cert formats for all data sources.

UI Updates​

  • Styling updates in pattern view
    • It is easier to differentiate between elements of different data source types.
    • A lexicon icon is presented on the element when a lexicon filter is configured to an event type.
    • Multiplicity is presented in a more readable and intuitive notation.
  • Event Type expanded details now contain all Event Patterns it is an input or output reference of.
  • The CLOSE button on the top right corner will now only close the immediate view, and not the entire project. A secondary button option is provide to close the project.
  • Link analysis and Manual Actions no longer available to non-Kafka event types.

Other​

  • Standardize styling on confirmation dialogues e.g. delete buttons are now always red
  • License error and permission error splash pages are now standardized across both Workstation and Authoring.

Workstation​

AI Configs​

  • AI Configurations now support Amazon Bedrock inference profiles, giving you greater flexibility to select the models that best fit your performance, cost, and capability requirements.

Chatbot​

  • The Workstation Chatbot can now return markdown including Github Flavored Markdown for better readability and formatting of responses.
    • TIP: In the AI Config System Prompt, request that the LLM responds in Markdown.
  • The chat session slides is now hidden for event-initiated chats.

Custom Fields​

  • Collection Custom Fields can now be automatically populated when Events are added, saving analysts time and eliminating common mistakes.

Interactive Map Widget​

  • Heatmap Mode: A new heatmap visualization is now available, providing a density-based view of geographic event data.
  • Clustering Control: Clustering can now be toggled on or off to suit your preferred map view.
  • Layer Visibility: Points and shapes can now be selectively shown or hidden, giving you greater control over what's displayed on the map.
  • Field-Level Visibility: Geo fields on an event can now be individually toggled, including the ability to hide all fields for a given event type at once.
  • Custom Color Assignment: Colors can now be set ad hoc for individual geo fields or all geo fields across an event type, making it easier to visually distinguish data on the map.
  • Deprecations: The 3D globe, hex bar visualizations, and cluster sizing slider has been removed.

File Uploads​

  • Users uploading files now receive clear error indicators that distinguish between files that cannot be uploaded and files that will be automatically renamed — reducing confusion and preventing unexpected upload results.