# Canakkale 5.4.0 ## Release Summary Autom Mate **Canakkale 5.4.0** introduces the **General Availability (GA) release of AI Agent Composer**, transitioning the capability from beta into a fully supported production feature. AI Agent Composer enables users to build intelligent agents capable of orchestrating Autom flows, invoking tools dynamically, and interacting with enterprise systems through AI-driven workflows. With this release, the platform introduces a production-ready infrastructure for agent-driven automation, including streaming response delivery, tool-calling loop execution, and expanded provider support. In addition to the AI Agent Composer GA launch, this release includes improvements across observability, automation debugging visibility, scheduling controls, and user interface consistency. Several stability issues affecting Playground streaming behavior, automation execution visibility, RESTful table operations, and large file handling have also been resolved. Overall, Canakkale 5.4.0 expands the platform’s capabilities toward **AI-powered automation orchestration** while improving system reliability and operational transparency. *** ## **Highlights of this Release** ### 🅱️ **Bug Fixes** * Fixed RESTful API table operations incorrectly inserting rows. * Fixed duplicate variable name validation missing in variable definitions. * Resolved Autom loading white screen issue. * Fixed duplicate “Unknown Error” messages in Playground. * Fixed AI agent creation issue where role prompts were shared between agents. * Fixed extremely slow typing effect during Playground streaming responses. * Fixed Mate Chat channel configuration modal validation behavior. * Fixed operation logs remaining stuck in “Processing logs” state. * Fixed Out of Memory issue when processing large file transfers. * Fixed MateAgent desktop capability restrictions not applied correctly. * Fixed AI Agent flows failing due to Autom structure changes. *** ### ✨ **New Features** * **AI Agent Composer is now Generally Available (GA).** * Streaming architecture for AI response generation. * Server-Sent Events (SSE) streaming infrastructure. * Tool-calling loop capability for AI agents. * Azure OpenAI provider support. * Observability metrics for job execution monitoring. * Mate Agent enriched payload generation endpoint. * HTML email template for automation summary reports. * Inspector insights for library usage and Autom structure. *** ### 🛠️ **Enhancements** * Autom error messages now display the **failing action name**. * Autom list now supports **schedule start/stop controls directly in the interface**. * Schedule configuration redesigned with **daily time configuration**. * Playground interface redesigned for improved usability. * Logs module messaging and reliability improvements. * Autom execution details can now be viewed directly after run completion. * Updated header layout including improved Import/Export placement. * Modal interaction pattern updated to use **Save instead of Edit**. * Vault custom credential creation UI improved. * Time Saved widget metric display updated. * Default mail server configuration can now be stored for client environments. * Azure OpenAI connector language configuration fields added. *** ## **Detailed Version** {% tabs %} {% tab title="🅱️ Bug Fix" %} * #### RESTful API Table Row Insertion Issue RESTful API table operations could insert rows incorrectly under certain data manipulation scenarios. The insertion logic has been corrected to ensure row operations maintain consistent table structures. *** * #### Duplicate Variable Name Validation Variables could previously be created with duplicate names in certain scenarios. Validation has been added to ensure variable names remain unique within their scope. *** * #### Autom White Screen Rendering Issue In some situations Autom pages could load as a blank white screen due to UI state inconsistencies. Rendering behavior has been stabilized to prevent page initialization failures. *** * #### Playground Duplicate Error Message Playground could display the same “Unknown Error” message multiple times. Error messaging logic has been corrected to ensure a single clear error message is shown. *** * #### Agent Role Prompt Sharing Issue Role prompts configured for one agent could appear in other agents due to configuration sharing. Agent prompt configuration is now isolated per agent. *** * #### Playground Streaming Typing Delay Streaming responses in Playground could render extremely slowly for long outputs. Rendering performance has been optimized to maintain consistent typing speed. *** * #### Mate Chat Channel Configuration Validation Configuration validation inside the Mate Chat channel modal did not always trigger correctly. Validation behavior has been corrected to ensure configuration errors are properly detected. *** * #### Operation Logs Stuck in Processing State Operation logs could remain indefinitely in a “Processing logs” state even after execution completed. Log state transitions have been corrected to reflect execution completion accurately. *** * #### Large File Transfer Memory Issue Execution Large file transfers through automation flows could trigger Out-of-Memory conditions. Memory handling during file transfers has been optimized to improve stability. *** * #### Desktop Action Capability Restrictions Desktop action capability restrictions based on MateAgent configuration were not consistently enforced. Restriction validation has been corrected to ensure actions follow MateAgent capability settings. *** * #### AI Agent Flow Compatibility with Autom Structure Recent Autom structural updates caused compatibility issues with certain AI Agent flows. Flow adaptation logic has been updated to ensure proper execution with the current Autom structure. {% endtab %} {% tab title="✨ New Features" %} * #### AI Agent Composer General Availability AI Agent Composer is now generally available and fully supported for production environments. The GA release introduces a stable platform for designing AI-powered agents capable of orchestrating automation workflows, invoking tools dynamically, and interacting with enterprise systems through Autom flows. *** * #### AI Streaming Architecture A streaming architecture has been introduced for AI responses. This enables AI responses to be delivered progressively rather than waiting for a complete response, improving user interaction experience. *** * #### Server-Sent Events (SSE) Streaming Server-Sent Events are now used to deliver streaming AI responses from backend services to the user interface. This mechanism enables reliable real-time communication for AI-driven workflows. *** * #### Tool Calling Loop for AI Agents AI agents can now iteratively call tools until a task objective is completed. This allows agents to dynamically interact with automation workflows and external systems. *** * #### Azure OpenAI Provider Support Azure OpenAI can now be used as an AI provider for AI Agent Composer and AI Agent integrations. *** * #### Mate Agent Enriched Payload Endpoint A new endpoint allows Mate Agent to generate enriched payload structures used in AI agent workflows. *** * #### Observability Job Execution Metrics New observability capabilities provide metrics and improved visibility into job execution performance. *** * #### HTML Email Template for Automation Reports A standardized HTML email template has been introduced for automation summary reports. *** * #### Inspector Library Insights Inspector now provides insights into library usage and dependencies within Autom flows. {% endtab %} {% tab title="🛠️ Enhancements" %} * #### Failing Action Display in Error Messages Autom error messages now display the specific action that caused the failure, improving debugging clarity. *** * #### Autom List Schedule Controls Users can now start or stop scheduled automations directly from the Autom list view. *** * #### Schedule Configuration Redesign Schedule configuration has been redesigned to use daily execution time settings. *** * #### Playground Interface Redesign Playground UI has been updated to provide a more compact and modern interface. *** * #### Logs Module Improvements The Logs module received improvements to messaging clarity and functional reliability. *** * #### Autom Execution Details After Run Execution details are now accessible immediately after an Autom finishes running. *** * #### Header Import/Export Improvements Import and export actions have been reorganized in the header for improved usability. *** * #### Modal Interaction Improvements Modals now use Save actions instead of Edit buttons to clarify interaction behavior. *** * #### Vault Custom Credential UI Update The Vault custom credential creation interface has been visually improved. *** * #### Time Saved Widget Metric Update The Time Saved widget now displays metrics using updated units *** * #### Default Mail Server Configuration Default mail server settings can now be stored for client environments. *** * #### Azure OpenAI Connector Language Fields Additional language configuration fields have been added to the Azure OpenAI connector. {% endtab %} {% endtabs %} *** ## User Notices & Guidance This section highlights important behavioral changes, usage recommendations, and practical tips introduced in Canakkale 5.4.0. Reviewing these notes will help you adapt existing workflows smoothly and ensure optimal system usage. workflows smoothly and make the most of the new versioning and storage capabilities. {% hint style="danger" %} #### Warnings (Critical Behavior Changes) * **AI Agent Composer Production Availability**\ AI Agent Composer is now generally available and intended for production use.\ Organizations previously testing the feature during beta should review agent configurations and provider settings before deploying agents in production environments. * **Schedule Configuration Update**\ Schedules now use daily execution time instead of start date configuration. * **Variable Name Validation Enforcement**\ Duplicate variable names are no longer allowed and must be unique within their scope. {% endhint %} *** {% hint style="info" %} #### Hints (Productivity Tips) * AI Agent Composer can orchestrate Autom flows using **tool-calling loops**, enabling agents to execute automation steps dynamically. * Streaming responses provide **real-time visibility into AI reasoning and responses** in both Playground and Composer environments. * Azure OpenAI support allows organizations to integrate **enterprise-managed AI models** into agent workflows. * Observability metrics can help track **AI-driven automation job performance** and execution behavior. {% endhint %} *** {% hint style="success" %} #### Best Practices (Recommended Usage) * Perform development and testing activities in Developer versions before releasing changes. * Regularly monitor complex flow chains that rely on error handling and Go To transitions. * Keep credential management centralized in Vault to benefit from automatic synchronization. * Leverage Monitoring hierarchy improvements as part of routine operational validation. {% endhint %} --- # Agent Instructions: Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter: ``` GET https://docs.autommate.com/userguide/release-notes/canakkale-edition/canakkale-5.4.0.md?ask= ``` The question should be specific, self-contained, and written in natural language. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.