UiPath announced updated and enhanced capabilities across the UiPath Platform. Enterprise automation programmes typically stall for three reasons: building new automations runs through a queue of developer capacity, legacy applications and on-prem databases sit outside what the platform can reach, and as agents make more decisions on their own, security and compliance teams need proof of what those agents did rather than a description of what they were supposed to do.
The new capabilities strengthen the platform’s control plane across governance, data, connectivity, and deployment, so agents, processes, and workflows stay aligned to business outcomes and inside enterprise-defined boundaries.
“Our platform gives customers the map they need to chart the right course for agentic transformation, at the pace and context their enterprise requires,” said Raghu Malpani, Chief Product and Technology Officer, UiPath. “That context informs the control plane that governs building, data access, integration, and the launch of automations that run the business, with the assurance that it all stays inside the boundaries the business sets.”
Build and test faster with UiPath Coding Agents and UiPath Delegate
UiPath Coding Agents, announced in May 2026, are now generally available on the UiPath Platform, enabling teams to discover and build automations, then optimise them across the full lifecycle. Enterprises can use UiPath Coding Agents, Claude Code, Cursor, Codex, Antigravity, or other supported agents, each natively fluent in UiPath folders, jobs, queues, assets, audit logs, and connections through natural-language requests against a live, logged-in session, with no manual API lookups required.
UiPath Delegate is a new productivity agent built for the enterprise that acts on a user’s behalf to complete administrative and role-specific tasks, including the operational work around software testing. Delegate helps testers gather context, move information between systems, update records, attach evidence, and prepare summaries. It can save repeatable workflows from tester actions and recorded sessions as reusable routines, then run them on demand, on a schedule, or from defined triggers, while testers retain control of judgment, strategy, quality decisions, and final sign-off.
UiPath is also extending coding agents into UiPath IXP to automate the manual setup work behind document extraction projects. A coding agent creates the project, checks its predictions against source documents, tunes low-scoring fields, and publishes the result, sharpening its instructions and examples with each pass, while a person sets the quality bar and approves the outcome.
Verified enterprise context with UiPath Data Fabric and UiPath Integration Service Two platform enhancements keep data secure, governed, and in the right context when building and running processes and workflows. UiPath Integration Service adds four capabilities:
- Database Hub – reads and writes to systems like SQL Server, Oracle, and Databricks without custom code
- Relay – connects cloud automations to on-prem and ground systems
- Third-Party Credential Store – lets customers use their existing enterprise credential store when creating connections, instead of duplicating credentials
- MCP Connector – lets workflows call tools from any compatible MCP server as a native activity, instead of a custom-built adapter
UiPath Data Fabric now turns distributed enterprise data into a governed, reusable context for agents, automations, apps, and workflows. Agents can query native entities directly as generally available context, grounding decisions in current business records instead of flat API responses. Federated Entities give zero-copy connectivity to external sources like Salesforce and SAP, with improved tooling, a new API, and a full audit trail, without copying the underlying data. Builders can use the same governed entities across the platform, including orchestration and coding agents, so data and automations ship and run together. Governance at scale New controls at every layer of the UiPath Platform keep models, actions, data, and permissions inside enterprise-defined boundaries as AI takes on more autonomous work. One policy framework applies across the board, so every actor doing the work, agent, robot, or person alike, answers to the same rules:
- Model Hub – gives teams central visibility into which models are in use, where they run, and how they’re routed
- Runtime Checker – validates agent behavior continuously against defined policies while an agent is running, not only when it was built
- LLM-as-Judge Guardrail – evaluates outputs against custom criteria
- Compliance Packs – map regulatory and internal standards into trackable controls
- Identity & Access Policies – govern who or what can reach a given model, tool, or resource
Automation Suite for Linux: the full stack, on premises
UiPath Automation Suite is now available on Linux with the full UiPath Platform agentic AI stack. Organisations with strict regulatory, sovereignty, or security requirements get full control and visibility over where data runs and how policies are enforced, keeping ownership of the environment in-house. Agents, robots, and people work together in a governed platform, keeping critical workflows, data, and controls inside infrastructure the enterprise already operates.
Discussion about this post