- Getting started
- Prerequisites
- Building agents in Studio Web
- Building agents in Agent Builder
Agents user guide
Escalations and Agent Memory
As a developer, you may want to give your agent a way to validate business case details with a human before running a tool, or to request manual assistance if they run into issues. In other cases, there may be business exceptions in a process when you want an agent to have a human make an approval or some other type of decision.
This is where escalations come in. Escalations are powered by Action apps in Action Center and give developers the tools to design and configure human-in-the-loop events in agent execution.
Agent Memory is a service inside each agent that helps it remember facts and observations as it works. The service stores escalations for each agent run at design time, and for running processes with agents at runtime, to support a long-term memory the agent can refer to, in addition to Context Grounding indexes.
Each resolved escalation in Action Center stores a key-value pair of the question and supporting context. These are then matched using semantic similarity to incoming queries. The question (key) has an 80 character limit, and the key-value pair has a TTL (time-to-live) of 3 months.
Enable Agent Memory in the Escalations panel. Once enabled, Agent Memory is tied to the latest version of the published agent. At runtime, this version stores by default store each Action Center task triggered by an agent escalation. Users resolving the escalation also have the option to not store the escalation in agent memory.
You can create multiple escalations for different situations where an agent may require human assistance.
Use the Properties panel to add and configure the escalation.
- Select an action app form your available resources. You must first deploy an app to your tenant.
- Add a Prompt to help the agent know when to use the escalation.
- Optionally, enable Agent Memory. Activate this option to store question and response pairs from escalations, and use text embeddings to automatically resolve escalations with content that is very similar to past escalations. Agent Memory stores design time and runtime escalations separately.
- In the Assignments section, add the escalation Recipient. You can assign an escalation to a user in your UiPath® organization.
- In the Inputs section, see the list of fields that are marked as required in the selected app. Describe the content of the field, so that the agent can infer its value and fill it in when raising the escalation.
- In the Outcome behavior section, specify how each outcome should be handled by the agent after an escalation was resolved.
After you create an escalation, you can reference it from the system prompt. Use explicit instructions, such as: "Before using Tool X, be sure to raise the Human Confirmation escalation and include the details for the action you're proposing".
Agents do not currently support the In/Out argument type for Action Schema properties.
