Automic Automation V26: The Top 7 Features That Redefine Enterprise Workload Orchestration

Automic Automation V26 delivers a governed, vendor-agnostic environment for orchestrating Agentic AI and modern DataOps workloads across hybrid cloud infrastructures. It removes long-standing technical barriers by placing an Intelligent Control Plane at the heart of the platform, enabling both expert automation engineers and a broader group of non-expert automation builders to design and execute operational work within a policy-bounded framework.

In this article, we at Tricise — long-standing Automic Automation specialists — break down the seven most impactful enhancements in Automic Automation V26 that your automation team should know about, starting with the single most important one: AI Jobs.

1. AI Jobs and AI Connection Objects: The Foundation of Agentic Automation

The headline feature of Automic Automation V26 is the introduction of AI Jobs and the accompanying AI Connection Objects. Together, they represent a shift from static, scripted automation toward dynamic, self-orchestrating systems. Automic Automation V26 delivers the infrastructure to build and deploy AI Agents — intelligent entities capable of interpreting complex instructions, interacting with external systems via MCP, and executing decision-making tasks within workflows.

What Are AI Jobs?

An AI Job is a new object type that enables the direct integration of Large Language Models (LLMs) and Model Context Protocol (MCP) servers into automated workflows. AI Jobs extend the traditional Job concept, allowing you to define AI-driven tasks that previously required human intervention.

The capabilities of AI Jobs include:

• Automating complex decisions: AI Jobs can analyze scenarios and then suggest or autonomously execute remediation steps.
• Actionable insights: They can summarize reports, logs, or external data sources and pass structured insights to subsequent workflow steps.
• Auditability and security: Every action is logged against policies defined in the connection object, providing an audit trail for compliance.
• Object-oriented design: While the earlier ASK_AI script function is still suitable for simple, one-off text queries, AI Jobs maintain conversation history, can use external tools via MCP, and keep context across multiple workflow steps.
• Natural-language configuration: Business users can configure AI Jobs using natural-language prompts rather than complex coding.

AI Connection Objects: Governance Where It Matters

Every AI Job must have an AI Connection Object assigned to it. These objects serve as the central infrastructure bridge between the Automation Engine and the Automation.AI component. Two design points stand out:

1. Central model management: Administrators define LLM and MCP server configurations once, rather than in every individual AI Job. For example, changing the underlying model (upgrading from GPT-3.5 to GPT-4o) is a single change in one place, not across hundreds of jobs.
2. Scoped access by team: A Finance AI connection can expose CSV tools, while an IT Ops AI connection exposes Automic Automation tools. Teams get precisely the capabilities they are entitled to.

Importantly, AI Connection Objects do not store authentication credentials or API keys. These sensitive tokens are managed inside the Automation.AI component and are never exposed in the Automic Web Interface or in Transport Cases — a critical design point for any regulated enterprise.

Net-net: AI Jobs are the building blocks of agentic workflows in Automic V26, and AI Connection Objects are the governance layer that makes them enterprise-ready.

2. AI-Augmented Workflow Creation: Welcome to “Vibe Coding”

Automic Automation V26 introduces Augmented Workflow Creation, also referred to as Text-to-Flow or, more playfully, Vibe Coding. This capability uses generative AI to turn natural-language conversations into production-ready workflows.

The Automation Assistant engages the user in a two-way dialog, clarifies requirements, and then generates:

• A detailed work plan for approval before anything is created
• Production-ready objects such as Workflows, Jobs, Connection objects and Login objects
• Suggested folder structures and orchestration logic
• Built-in failure handling

The user can iterate conversationally until the implementation matches the business requirement. This feature is a clear driver of “citizen developer” adoption — empowering line-of-business staff to automate tasks without deep knowledge of AE scripting or Automic-specific concepts.

Important limitation: The AI Augmented Workflow Creation feature is currently supported only by the Google Gemini LLM. If you plan to use this capability, you need to select Gemini as your provider during configuration.

3. MCP Integration: A Standardized Bridge to the Generative AI Ecosystem

The Model Context Protocol (MCP) integration is one of the most important architectural additions in Automic Automation V26. MCP functions as a universal interface between Automic Automation and the generative AI ecosystem. Through the Automation.AI component, the system dynamically discovers and maps capabilities from MCP servers — such as database connectors, API wrappers, or local files — directly into AI Jobs.

The documented business benefits include:

• Standardized communication: MCP acts as a universal translator that converts automation needs into precise, reliable AI instructions regardless of the underlying model.
• No custom “glue code”: New data sources and tools can be plugged in via MCP without bespoke integration work per LLM.
• Reduced MTTR: Operators can diagnose failures using natural-language queries instead of manual log digging.
• Enhanced security: The Automic MCP Server now supports OAuth 2.0 and Basic Authentication, with separate Authentication In and Authentication Out configurations. Authentication In validates the identity of clients calling the MCP Server, including OAuth 2.0 via a JWKS URL (supporting providers like Microsoft Entra ID and Okta). Authentication Out controls how the MCP Server handles authorization headers before forwarding requests — pass-through of the client’s original token, enforced token types (Bearer or Basic), or injection of an encrypted, hardcoded Basic auth header for service accounts.
• Ecosystem flexibility: The same integration model covers custom on-premises LLMs, Broadcom’s Gemini offering in Automic SaaS, and — new in V26 — Azure OpenAI and VMware PAIS (Private AI Services).

For teams already running hybrid environments, this is a significant simplification. See the official Broadcom MCP integration documentation for the technical deep dive.

4. The AI-Powered Code Assistant: Script Faster, Script Better

Scripting has traditionally been one of the most time-consuming aspects of Automic development. Automic Automation V26 addresses this with the AI-powered Code Assistant, available inside the script editor of the Automic Web Interface.

Using the new Ask Intelligent Assistant function, users submit a prompt in natural language describing what the script should do. The assistant inserts the relevant code snippet inline and explains its logic in the Intelligent Assistant panel. The assistant can generate code in:

• Python
• Windows BAT and PowerShell
• Linux Shell

Beyond generation, the Code Assistant also performs:

• Code optimization: Select a block, ask for improvements, receive a more efficient snippet.
• Code replacement: Mark lines, ask the assistant to rewrite them, and the result replaces the selection.
• Inline interaction: Interact directly within the script editor; the generated code is inserted on the line below.
• Contextual awareness: The assistant maintains context across the edit session of a script.

Two new script elements — BEGIN_AI_CHAT and END_AI_CHAT — round out the scripting experience by letting scripts maintain persistent conversational context across multiple ASK_AI calls via a new ChatID parameter. Scripts using ASK_AI without ChatID continue to work unchanged; the parameter is purely additive for workflows like dynamic error handling or iterative code generation.

V26 also adds a range of new string and text processing script functions with full Unicode support and regular-expression capabilities — useful complements for teams working with unstructured data inside AE scripts. For organizations investing in internal automation skills development, the combination of Code Assistant and richer scripting primitives substantially lowers the learning curve.

5. Enhanced Python Integration: First-Class Support for Data Engineers

The native Python Job types — Windows Python, Unix Python and Generic Python — originally shipped in version 24.5 for SaaS customers only. Automic Automation V26 makes them available to on-premises customers as well.

Key highlights:

• Direct Python execution: Run Python code natively in Automic Jobs and use any Python library by specifying it in the job definition.
• New PIP report type: Logs the output of the Python package manager when installing the packages specified in the Job, simplifying dependency troubleshooting.
• Python dictionary of Automic variables: The _automic_variables dictionary exposes all Automic Automation variables to Python, enabling bidirectional integration without AE scripting.
• Array support (new in V26): Automic arrays are now accessible as Python arrays, and Python arrays can be registered back as Automic variables via the automic_register_variable function. A typical use case: a Python Job retrieves firewall rules from GCP, stores them in a Python array, and registers that array back as an Automic variable for downstream jobs.

For teams running DataOps pipelines that combine Automic orchestration with Python processing, this is a meaningful productivity improvement.

6. Workflow Editor and Monitor: More Visibility, Less Clicking

Automic Automation V26 introduces a substantial set of visual and usability improvements to the Workflow Editor and Workflow Monitor. These changes are immediately visible to daily operators and developers.

Task Property Icons at a Glance

As soon as a task property is configured and saved, a dedicated status icon appears dynamically on the task box. Hovering over the box reveals all configured properties in a tooltip. The exceptions: no icon is displayed for Variables, PromptSets or Runtime settings.

Expanded Filtering

The Workflow Editor adds three new filter criteria: Breakpoint, Calendar and Task Time (Earliest Start Time and Time Checkpoints).

The Workflow Monitor goes further, adding filters for:

• Operational States: Task Status, Active/Inactive, Breakpoints
• Logic and Conditions: Pre/Post-Conditions, Calendars, Rollback settings
• Execution Details: Comments, Earliest Start Time, Time Checkpoint

Five-Color Traffic Lights and Per-Task Progress Bars

Workflow Monitor traffic lights have expanded from four to five statuses — red (abended), amber (blocked), grey (waiting), blue (active), and green (successfully ended). Selecting a traffic light highlights the tasks in that status. Each individual task box now carries its own progress bar, providing a granular view of execution status at the task level.

Improved Zoom Behavior

Zoom thresholds have been recalibrated: the “Full Card” view persists across a wider zoom range, and task containers transition more logically between simplified shapes and full-detail cards. Time-related labels in task boxes have also been shortened (for example, Estimated Start Time becomes Start (EST)) to improve readability.

Time and Dependencies, Finally Separated

The former Time & Dependencies properties tab, long considered cluttered, has been split into two cleanly separated tabs: Time and Dependencies.

7. Modernized AWI, New Status Colors and Stronger Cloud Integrations

The Automic Web Interface (AWI) has received a visual refresh in Automic Automation V26 with a refined color palette, a redesigned icon set, and stricter accessibility standards.

Key visual and usability enhancements:

• Connection color on the entire top menu bar: Instead of a thin line, the selected connection color now fills the header bar, making it immediately obvious whether you’re in a production client, a development client, or Client 0 — a useful safety feature for any organization running multiple Automic environments.
• 25 curated connection colors: Accessibility-compliant, with hardcoded text and icon contrast. Custom free-form AWI theme colors and custom connection colors have been deprecated to prevent unreadable UIs.
• Refreshed task status colors: Applied consistently across the Workflow Monitor, task lists, dashboards and widgets to support faster at-a-glance incident response.
• Modern browser support required: The new AWI requires Chrome 138+, Edge Chromium 138+, Firefox 140+, or Safari 18.5+.

Cloud Integrations: A Hybrid-Cloud Orchestrator for the Enterprise

One of the less glamorous but strategically decisive shifts in Automic Automation V26 is the consolidation of cloud integrations onto the modern agent framework originally introduced in V24. For teams running workloads across multiple hyperscalers — which, realistically, is most large enterprises today — this means consistent behavior, shared observability, and a single security model across every cloud touchpoint. The practical impact: you stop maintaining a patchwork of legacy connectors and start treating AWS, Azure, GCP, and SAP workloads as first-class citizens inside the same workflow.

Here is what the updated integration matrix actually delivers, grouped by where your workloads run:

AWS

The AWS Lambda agent lets you trigger and orchestrate serverless functions directly from Automic workflows — useful for event-driven data processing, on-demand compute spikes, and lightweight integration glue between enterprise systems and cloud-native services.

Microsoft Azure

V26 ships the broadest Azure integration portfolio to date. Azure Functions covers serverless execution, Azure Logic Apps brings low-code workflow automation into Automic’s orchestration layer, and Azure Data Factory plus Azure Blob Storage handle large-scale data movement and staging scenarios. On the analytics side, Azure PowerBI enables automated report refresh and dataset operations — closing the loop between data pipelines and business reporting. Together, these cover the full Azure data-and-integration stack that most enterprises actually use in production.

Google Cloud Platform

The GCP portfolio in V26 is built around data and batch processing. Google Dataproc handles Spark and Hadoop workloads, Google Cloud Batch orchestrates large-scale batch jobs, Google Cloud Run brings containerized services into workflows, and GCP BigQuery integrates directly with Automic for data warehouse operations. For organizations building analytics or ML pipelines on GCP, this means Automic can drive the entire pipeline end-to-end.

SAP

SAP remains a core Automic stronghold, and Automic Automation V26 doubles down. SAP S/4 HANA Application Jobs, SAP Integration Suite, and SAP BOBI are all enhanced in this release. Combined with the new SAP Agent Multi-System Support (see below), which lets a single agent connect to multiple SAP target systems from the Job level, this makes Automic Automation V26 the strongest SAP-orchestration release Automic has shipped.

Workflow & Data Platforms

The Airflow agent lets Automic orchestrate Apache Airflow DAGs — a practical bridge for data teams that have standardized on Airflow for pipeline authoring but need enterprise-grade scheduling, SLA management, and governance on top. The REST Agent rounds out the picture with additional comparison operators in the Retry Settings of the Response page, making it easier to build resilient integrations against any REST-based service that does not yet have a dedicated agent.

The common thread across all of these: you no longer pick between “cloud-native orchestration” and “enterprise workload automation.” With Automic Automation V26, Automic becomes the single control plane that spans both.

Additional High-Impact Enhancements You Shouldn’t Miss

Beyond the top 7, Automic Automation V26 ships several enterprise-grade improvements that may be decisive for your specific environment.

Zero Downtime Upgrade (ZDU) with Version-Independent AWI

Automic Automation V26 allows an AWI version higher than the AE server version to be used to start and complete a ZDU upgrade or rollback — meaning you can upgrade AWI first and then drive the ZDU from there. This version mismatch is permitted only during a ZDU; for all other scenarios, AWI and AE must be on the same release. During the upgrade window, AWI displays a version-mismatch banner. A new ZDU_PORT parameter in the AE INI file lets administrators define a dedicated port for the REST process during ZDU (a random port is used if not set).

Database Maintenance Through AWI and the AE REST API

Automic Automation V26 integrates core AE DB Utility functions directly into the AWI Administration perspective and the AE REST API. Administrators can now perform archiving, revision report creation, reorganization, deletion of operational data, and emptying the Recycle Bin from AWI or via REST endpoints — without direct database access or local utility installations. The existing AE DB Utilities (Archive, Revision Report, Reorg, Unload) remain fully supported. Results are tracked in the Client History view and available for download.

Comprehensive OAuth 2.0 Across the Stack

Automic Automation V26 broadens OAuth 2.0 support significantly:

• XOAuth2 in SMTP: A new SMTP_AUTH_METHOD option in UC_CLIENT_SETTINGS and UC_SMTP_MYSERVER adds XOAUTH2 alongside PLAIN and LOGIN. A new SECRET type in the Login object supports encrypted storage of XOAuth2 credentials.
• UC_OAUTH_SETTINGS for the AE REST API: A new system variable enabling JWT-based OAuth2 security for the AE REST API, with JWKS endpoint caching, claim-based identification (email for users, appid for applications by default), and configurable JWT scopes.
• MCP Server OAuth 2.0: Covered in detail in section 3 above.

Enforced SSO-Only Login (MFA Compliance)

A new sso.enforced setting in AWI’s configuration.properties disables basic username/password authentication when SAML or Kerberos is enabled. This closes a common security gap where users could inadvertently bypass mandatory MFA by falling back to basic authentication.

Automatic Password Encryption

V26 adds an opt-in automatic password encryption feature for most configuration files — INI files (AE, Utilities, Java-based Agents, Analytics, TLS Gateway, Proxy, LDAP Sync, Automation.AI) and properties files (Analytics). The feature is disabled by default. When enabled, a unique 32-byte installation-specific key is generated and encrypted passwords are prefixed with --20. Automation.AI uses a separate, stronger pipeline: a FIPS 140-3 compliant combination of PBKDF2WithHmacSHA512 and AES-256-GCM, with encrypted values prefixed {cipher}.

Transport Case Enhancements

Three significant Transport Case changes in Automic Automation V26:

• Traceability: Every import/export now generates a detailed report stored in the AE database and surfaced in the Client History view, with errors and object lists for auditing.
• Sequential, resilient imports: The import processes objects one-by-one, allowing partial imports (skipping objects that fail and continuing with the rest), with a summary reported in the AWI import dialog.
• No more size limitations: The MAX_TC_SIZE system variable is deprecated. The export file is now called uc_data.txt (previously data.txt) for consistency with the AE DB Unload Utility.

SAP Agent Multi-System Support

Previously, the R3 Connection object had to be assigned at the SAP Agent level. In Automic Automation V26, the Connection object can be assigned at the Job level instead (for both SAP Jobs and Remote Task Manager Jobs), enabling a single SAP Agent to connect to multiple SAP target systems. Existing configurations are fully backward-compatible: a Job-level Connection takes precedence over an Agent-level one, and a missing assignment on both levels causes the Job to fail.

Containerized Agents (Linux and SAP)

Pre-configured, containerized agents can now be stored in any registry and deployed for either on-premises Automic or AAKE environments via standard Kubernetes and Helm charts. Containerized agents mandate TLS/SSL (non-TLS agents are not supported in this format) and run in non-privileged, restricted mode to comply with pod security policies. The current containerized offerings include Linux and SAP agents, plus a number of agent integrations.

Automatic Agent Authentication

Agents created through AWI, the REST API, or the AE.ApplicationInterface Java API using the LOCAL authentication method are now automatically authenticated, eliminating manual authentication steps. This applies only to agents created in V26 or later.

New AI REST Endpoints

Automic Automation V26 adds AI-specific endpoints to the AE REST API: GET /ai/llm (list available LLMs), GET /ai/mcp (list active MCP servers and their tools), and POST /prompt_ai (send a structured chat request to Automation.AI, aligned with the Google Generative Language API V2 format, supporting prompt_context, extra_tools, and mcp_filter for fine-grained control).

Granular CyberArk AppIDs

The Login object now supports per-entry CyberArk Application IDs via a new Metadata column, overriding the global AppID from UC_VAULT_CYBERARK where needed. Useful for organizations with multiple CyberArk vaults or strict per-application compliance requirements.

Customizable AI Prompts via UC_AI_PROMPTS

A new system-wide STATIC VARA object, UC_AI_PROMPTS, available in Client 0, stores the LLM query strings used for Report/Execution Analysis, Script Analysis, and Script Generation. Administrators can now fine-tune these prompts to match their LLM and environment — including changing the LLM’s response language. Read access is maintained even with Object Access Control Lists active.

High Availability for Automation.AI

Automation.AI now supports high availability. In AAKE, you configure this via the automationAiReplicas parameter in values.yaml. On-premises, you install multiple Automation.AI instances. In both cases, the instances share the same Automation.AI database (a new dedicated PostgreSQL database for Automation.AI data).

Conclusion: Automic Automation V26 Is the AI-Ready Release

Automic Automation V26 is more than a point release. The combination of AI Jobs, AI Connection Objects, MCP integration, Augmented Workflow Creation and the AI-powered Code Assistant establishes Automic Automation V26 as a platform designed to operate autonomous agents inside mission-critical environments under policy-bounded governance.

At the same time, the non-AI improvements — modernized AWI, richer workflow visibility, native Python support, broader OAuth 2.0 coverage, enforced SSO, integrated database maintenance, Transport Case without size limits, SAP multi-system support, and containerized agents — make the upgrade worthwhile even for organizations not yet ready to adopt generative AI in production.

If your organization runs Automic today, Automic Automation V26 is the release to plan for. If you are evaluating workload orchestration platforms for the AI era, this is the version that should be on your shortlist.

Want to discuss your Automic Automation V26 upgrade? Get in touch with Tricise for a free consultation, or explore our resources and events to learn more about smart automation, ValueOps and network observability.