SVG.svg "Cognigy - conversational ai platform for enterprise"){kind=logo}
.png?width=60&height=60&name=AI%20Copilot%20logo%20(mega%20menu).png)
.png?width=600&height=600&name=Knowledge%20AI%20Feature%20image%20(2).png)
.png?width=29&height=29&name=twitter%20(1).png){kind=small}
The release of Cognigy.AI v4.91 marks a significant milestone with the public availability of Agentic AI. This update introduces advanced features previewed in the launch webinar, adding impactful enhancements like condition-based Tool execution and image file processing.
Additionally, we have released a range of product enablement materials to help you get started effectively.
Watch our Launch Webinar for a Complete Overview of Agentic AI
Control and Optimize Agent Actions with Condition-Based Logic
Overview
Previously, we have introduced how Cognigy allows you to combine and transition seamlessly between dynamic Agentic behavior and predefined, structured workflows, catering to diverse enterprise use cases.
With this release, we strengthen the synergy between adaptability and control by introducing the ability to combine LLM-powered reasoning with rule-based logic for Agent action execution.
How it Works
In Cognigy.AI, Tools represent the specialized capabilities of an AI Agent to perform its job successfully. Depending on real-time user inputs, AI Agents dynamically reason and select the required Tool for goal completion.
The latest update introduces the ability to define conditional access to these Tools. For example, you can configure a submit_claim Tool to be available to the AI Agent only if the customer has an active insurance policy in place.
Benefits
This feature provides an extra layer of governance and control in dynamic and non-deterministic AI environments. Limiting the tools available to AI agents can improve precision by improving focus and reducing complexity and potential sources of error.
Enhanced Multimodality with Image File Processing
Overview
AI Agents can now also analyze and interpret image-based input in real-time. This enhancement expands multimodal capabilities, allowing the Agent to extract data from images and integrate it into various workflows like document verification, incident reporting, product support, and value-added services.
How it Works
To enable this feature, activate the Process Images option in the AI Agent Node settings. Additionally, ensure that your chosen LLM has the vision capability, and the Endpoint has a configured File Storage to allow image uploads.
On par with this improvement, Cognigy Webchat also supports the preview of uploaded images, providing a smooth and intuitive user experience.
Benefit
In an era where consumers expect to engage with businesses on their terms, this capability enables you to deliver rich experiences that adapt to how customers prefer to communicate.
Agentic AI Enablement Materials
With the public release of Agentic AI, we have published a variety of enablement materials to give you a head start.
🎓 Academy Certification course📄Product Documentation
👩💻Help Center How-to Guides
Other Improvements
Cognigy.AI
- Added the capability to enforce login with Single Sign-on (SSO), enhancing security and simplifying access management
- Added on-the-fly resizing for uploaded images
- Updated the Project Creation page. You can now select from two wizards for initial agent creation: Agentic AI or Intent-driven AI.
- Removed the restriction on the maximum number of buttons for the Text with Buttons output type in Say Nodes
xApps
- Improved the content structure of messages sent from xApps in the conversation chat within Cognigy Live Agent to enhance readability for human agents
Webchat
- Improved the UI of Webchat v3 by displaying uploaded images as previews in the chat
- Updated Webchat v3 and its Demo Webchat to v3.8.1
For further information, check out our complete Release Notes here.
