agents

latest

false

Agents user guide

Last updated Oct 30, 2025

Managing UiPath agents

The Agents page in Automation Cloud^TM serves as your centralized registry and observability hub for all UiPath agents. Use it to monitor performance, test conversational behavior, manage shared resources, and maintain visibility across your deployed ecosystem.

From a single location, you can:

Monitor deployed agents and review performance metrics in real time.
Test conversational agents before and after deployment.
Investigate and resolve agent incidents or degraded health conditions.
Access and deploy agent templates from the UiPath Marketplace.
Manage Context Grounding indexes.

Terminology

An Agent instance is a single, active execution of a deployed UiPath agent—representing one runtime occurrence where the agent processes input, performs reasoning, executes tools, and produces output according to its defined logic and context.
Agent instance management represents the set of tasks related to monitoring, maintaining, improving, and terminating agents and agent jobs.

Key features

Here are the key features of the Agents page in Automation Cloud^TM:

Real-time agent monitoring.
Error handling and recovery.
Runtime agent coordination between agents and human operators.
Merging and consensus handling via decision-making gateways.
Performance analysis and optimization based on multiple metrics.
Governance and compliance checks ensuring adherence to enterprise rules.

Overview

The Overview tab provides a high-level summary of your agent ecosystem, consolidating operational insights, performance indicators, and consumption analytics into one centralized view. It acts as a command center for understanding how agents are performing across your organization and how efficiently they are consuming resources.

Performance insights

The tab aggregates runtime data across all deployed agents to help you monitor performance trends and identify optimization opportunities. Key metrics include:

Total jobs: Tracks the total number of completed agent executions, providing a measure of activity and adoption.
Success rate: Measures the percentage of successful agent runs to help you evaluate reliability and detect recurring errors or instability in agent configurations.
Average response time: Indicates how quickly your agents are able to complete tasks, allowing you to monitor latency and efficiency at scale.
Agent unit consumption: Quantifies total unit usage across all agents, enabling better forecasting, budgeting, and resource allocation.

Monitoring and feedback analytics

The tab also surfaces consumption and feedback trends across your environment:

Agent unit consumption reports identify the most active or costly agents, helping prioritize tuning or resource allocation.
Feedback summaries highlight which agents receive the most user interactions or performance ratings—helping teams focus on improving user experience and impact.

Filter and time-based analysis

Administrators can analyze data across various dimensions, such as agent type, deployment state, or time range, to identify anomalies and evaluate trends over defined operational periods (e.g., daily, weekly, or monthly views). These filters enable targeted analysis and streamline reporting.

Deployed agents

The Deployed agents tab provides a consolidated view of all agents currently active in production, along with real-time performance, health, and feedback metrics. It acts as the primary workspace for monitoring operational efficiency, diagnosing issues, and ensuring that deployed agents deliver consistent, high-quality outcomes.

Unified agent visibility

This view aggregates all production-ready agents across your tenant, allowing administrators and developers to quickly assess operational status and recent activity. Each agent card summarizes key operational metrics, including:

Success rate – The percentage of successful runs, helping you identify degraded or underperforming agents.
Average response time – A measure of execution speed and efficiency across recent jobs.
Feedback volume – The number of user or system evaluations collected for the agent.
Health indicators – Visual cues (e.g., Healthy, Degraded) reflecting overall performance trends and runtime stability.

Agents can be filtered by state, type, folder, or time range, enabling precise monitoring across business areas or deployment environments.

Operational insights and performance tracking

The tab provides contextual analytics that extend beyond simple runtime status. You can explore:

Active and completed jobs – Separate timelines track live executions versus completed runs, helping identify workload peaks and bottlenecks.
Error and latency patterns – Performance charts display execution errors, time-to-first-response, and latency percentiles, allowing teams to fine-tune responsiveness.
Feedback trends – Aggregated sentiment analysis (positive vs. negative) shows how agent quality evolves over time and which updates affect user satisfaction.

These insights help diagnose systemic issues, validate new deployments, and prioritize optimization efforts.

Usage and consumption analysis

The dashboard highlights how agents interact with their environments and how resources are consumed:

Agent unit usage – Tracks total consumption and identifies outliers in cost or resource utilization.
Top capabilities invoked – Lists the most frequently used actions (tools, activities, or indexes) across agent runs, helping teams understand behavior patterns and common dependencies.
Most active users and workloads – Reveals usage concentration, offering transparency into who or what processes drive the highest demand.

Detailed agent runtime view

Selecting an individual agent opens a detailed analytics workspace, combining:

A runtime diagram that maps the agent’s logic, tools, and memory sources.
Health score visualizations that assess prompt design, tool integration, and evaluation coverage.
Execution traces with granular visibility into every run (inputs, outputs, errors, and timing details), allowing developers to reconstruct and debug the agent’s reasoning process.

The agent instance view provides full observability into individual agent executions, offering a complete trace of decisions, actions, and performance metrics within a single run.

Understanding the trace view

The trace view combines a graphical representation of the agent’s logic with time-aligned execution data, allowing you to inspect the flow and outcomes of each step. The canvas shows nodes representing major agent components like models and tools, connected in sequence.

Once you're in the Trace view of an agent run, follow these steps to interact with and investigate the trace data:

Preview node information. Hover your mouse over any node on the canvas to see a tooltip that includes:
- The execution status (success, retry, or failure)
- Start and end timestamps
- A preview of the input and output data
This is helpful for getting a quick sense of what happened at each step without needing to open the full details.
View full node details. Select any node to open a detailed panel which includes:
- Complete input and output payloads (JSON)
- Execution logs or error messages (if applicable)
- Runtime metrics, such as latency and token usage
- Configuration parameters, if available
Navigate between related steps, to explore how information flowed through the agent:
- Select connected nodes to move upstream or downstream.
- This helps you trace how decisions, inputs, or outputs from one step influenced the next.
Use the Execution Trace panel. This panel shows a chronological list of all recorded actions during the agent run.
- Select a row in the trace log to highlight the corresponding node on the canvas.
- Similarly, selecting a node on the canvas automatically highlights the relevant row in the trace log.

Conversational agents

The Conversational agents tab provides a complete environment for managing, testing, and analyzing agents that interact with users through natural language dialogue. It brings together design-time testing, live chat evaluation, runtime insights, and feedback analytics to help teams continuously improve conversational performance and reliability.

Conversational agents can be tested directly within the chat interface. This embedded playground allows you to simulate real-world exchanges and evaluate how the agent handles intent recognition, turn-taking, and context continuity. You can:

Engage with the agent in real-time, sending natural language prompts and observing live responses.
Test prompts or craft custom messages to assess coverage across scenarios.
Evaluate how the agent applies tools, memory, and models during a conversation loop.

This testing environment helps validate conversational logic before deployment, ensuring accuracy and tone align with intended use cases.

The Runtime view provides end-to-end visibility into how conversational agents perform in production, capturing live sessions, completion rates, and user interactions. It tracks key operational metrics, such as response time, duration, and agent unit consumption, to ensure performance, scalability, and reliability across workloads.

The Trace view enables debugging and optimization by mapping the full reasoning and execution flow of an agent. You can inspect decisions, tool calls, governance checkpoints, and performance heatmaps to understand behavior, validate logic, and enhance overall conversational quality.

The Feedback view aggregates user and system evaluations into a central dashboard. It visualizes sentiment trends, highlights recurring issues, and links feedback to individual runs for deeper context. This enables iterative fine-tuning and continuous quality improvement.

Draft agents

The Draft agents tab contains unpublished or in-progress agents currently being developed.

Each draft is displayed as a card that includes:

The agent name (or "Untitled", if not yet renamed).
The last updated timestamp.
A quick link to reopen the draft in Studio Web for further editing.

Draft agents can be resumed, edited, or published once finalized.

Context Grounding indexes

The Context grounding indexes tab enables you to monitor the status of your indexes and track their ingestion progress at the tenant level.

The Index Monitoring section provides graphical insights into the following metrics:

Queries: Tracks the number of queries performed on your indexes over a selected time interval.
Index jobs: Displays a count of ingestion jobs. If no jobs have been run in the selected time period, a placeholder graphic is shown.

The Indexes table lists all available indexes in your tenant. For more granular information on indexes, select an index from the table. The individual view provides the following information:

The number of Queries during the selected interval.
The index Sync job history during the selected interval. Select an entry in the sync table for a detailed view of the ingested documents.