- Getting started
- Prerequisites
- Building agents in Studio Web
- Building agents in Agent Builder
Agents user guide
About agents
Agents are the next evolution in automation. Powered by technologies like large language models (LLMs), machine learning, and traditional enterprise automation, agents are designed to operate in dynamic, non-deterministic environments. They can plan, act, learn, and adapt—making them ideal for processes that require judgment, flexibility, and contextual awareness.
Unlike deterministic systems such as RPA robots, which follow structured logic and fixed rules, agents use a probabilistic approach to make decisions based on patterns and real-time data. This makes agents highly suited for automating unstructured, exception-heavy workflows where conditions and outcomes vary.
| Agents | Robots (RPA) | |
|---|---|---|
| Nature | Probabilistic, adaptive | Deterministic, rule-based |
| Skills | Planning, decision-making, learning, natural language | Structured execution, efficiency, reliability |
| Best for | Unstructured tasks, ambiguous environments | Repetitive tasks, structured workflows |
| Triggering | User requests or system events | Schedulers, triggers, or attended execution |
| Collaboration | Work with users, robots, and other agents | Work independently or under agent direction |
Agents are dynamic, context-aware systems that operate based on goals. They communicate in natural language, plan their steps, make independent decisions, and escalate to humans when needed. Robots bring stability, compliance, and precision. Together, they enable agentic automation—a new model of automation that blends intelligence with action.
Each agent consists of four core components:
-
Prompt: Natural language instructions that define the agent’s role, goal, and constraints. Prompts can be user-driven or system-generated.
-
Context: Information sources the agent uses to ground decisions—such as memory, knowledge bases, or previous interactions.
-
Tools: Actions the agent can take—invoking automations, using APIs, triggering microservices, or collaborating with other agents and robots.
-
Escalation paths: Human-in-the-loop mechanisms (like Action Center or messaging channels) that enable review, approval, or assistance when needed.
-
Communicative: Use natural language to collaborate with users and systems.
-
Initiating: Triggered by system events or user input.
-
Planning: Break down goals into executable steps.
-
Deciding: Make real-time decisions based on patterns and current state.
-
Adapting: Access and respond to live enterprise data.
-
Healing: Identify and recover from broken workflows.
-
Learning: Retain memory across sessions to improve over time.
-
Coordinating: Work alongside other agents, robots, and people.
Not all tasks are agent-friendly. Agents are best used for well-scoped tasks in environments that benefit from adaptability and learning. Tasks with high accuracy, legal, financial, or regulatory constraints should continue to rely on deterministic automation.
Good use cases for Agents
-
Drafting and summarizing content (emails, messages)
-
Multi-system research and data gathering
-
First-pass customer interactions or ticket triage
-
Orchestrating narrow AI agents into larger workflows
Poor use cases for Agents
-
High-risk financial transactions
-
Regulatory workflows with zero error tolerance
-
Complex multi-step data processing without deterministic safeguards
Agentic automation is the orchestration of agents, robots, and humans in a unified automation ecosystem. On the UiPath Platform™, agents can operate with built-in guardrails, governance, and security—making them enterprise-ready from day one.
This architecture enables businesses to:
-
Combine LLMs and AI agents with structured RPA for broader automation coverage.
-
Automate processes across CRM, ERP, and other systems.
-
Use real-time data to power decisions and outcomes.
-
Expand automation into previously unreachable, unstructured areas.
Learn how to elevate your processes end-to-end with agentic orchestration using Maestro.
The following conceptual diagram presents the UiPath agentic automation loop. It shows how agents operate within the UiPath ecosystem, integrating memory, tools, context, and human input to deliver business outcomes. This architecture is evolving. Expect updates as capabilities expand.
