US-12619488-B1 - Agent health score for agentic automations

Abstract

A method is provided. The method is performed by a hyper-automation system performing agentic automations. The hyper-automation system include a memory storing for the agentic automations and for agent health score generation. The hyper-automation system include at least one processor executing the computer code to cause the method to optimize the agentic automations. The method includes generating a set of standardized metrics for application to an agent executing agentic automations. The method includes performing an evaluation of the agent executing the agentic automations that determines performance metrics by measuring success rates, error frequencies, and execution efficiency for different use cases or business processes. The method includes comparing the performance metrics across different use cases or business processes for the agent. The method includes generating agent health scores based on comparing the performance metrics and agent health score definitions.

Inventors

• Chibi VIKRAMATHITHAN
• Dragos H. Bobolea
• Jordan HOLCOMBE
• Zach Eslami
• Venkata Syam P. RAPAKA
• Emanuela HALLER
• Andrei Dumitrescu
• Andrei RUSU

Assignees

• UiPath, Inc.

Dates

Publication Date

20260505

Application Date

20250522

Priority Date

20250410

Claims (19)

1. 1 . A method of a hyper-automation system performing one or more agentic automations, the hyper-automation system comprising a memory storing for the one or more agentic automations and for agent health score generation, and the hyper-automation system comprising at least one processor executing the computer code to cause the method to optimize the one or more agentic automations, the method comprising: generating a set of standardized metrics for application to an agent executing one or more agentic automations; performing, during runtime, an evaluation of the agent executing the one or more agentic automations that determines performance metrics by measuring success rates, error frequencies, and execution efficiency for two or more different use cases or business processes; comparing the performance metrics across the two or more different use cases or business processes for the agent; generating one or more agent health scores comprising a metric based on comparing the performance metrics and one or more agent health score definitions that identifies an execution performance of the agent; and executing the agent with respect to the runtime based on a confirmation of the one or more agent health scores of the execution performance of the agent.
2. 2 . The method of claim 1 , wherein the hyper-automation system identifies the agent for the agent health score generation.
3. 3 . The method of claim 1 , wherein the hyper-automation system identifies an action or a task within a workflow of the one or more agentic automations assigned to the agent.
4. 4 . The method of claim 1 , wherein the one or more agent health scores comprise a metric measuring agent performance as the agent executes the one or more agentic automations.
5. 5 . The method of claim 1 , wherein the hyper-automation system generates the one or more agent health score definitions for measuring the success rates, the error frequencies, and the execution efficiency.
6. 6 . The method of claim 5 , wherein the one or more agent health score definitions comprise evaluation infrastructure, runtime performance, developer experience, security and compliance, and business impact.
7. 7 . The method of claim 1 , wherein the performance metrics align with organizational priorities that provide customizable weightings.
8. 8 . The method of claim 7 , wherein the customizable weightings comprise a scale from zero (0) to one hundred (100) with a scoring criteria represented as a percentage.
9. 9 . The method of claim 1 , wherein the execution performance comprises latency and average response time, reliability and error rate, compliance and throughput.
10. 10 . The method of claim 1 , wherein the hyper-automation system utilizes the metric to identify errors of the agent and improve performance of the agent.
11. 11 . The method of claim 1 , wherein the confirmation of the one or more agent health scores of the execution performance of the agent comprises receiving a selection of an interface icon for good performance.
12. 12 . A hyper-automation system performing one or more agentic automations, the hyper-automation system comprising: a memory storing for the one or more agentic automations and for agent health score generation; and at least one processor executing the computer code to cause the hyper-automation system to optimize the one or more agentic automations by: generating a set of standardized metrics for application to an agent executing one or more agentic automations; performing, during runtime, an evaluation of the agent executing the one or more agentic automations that determines performance metrics by measuring success rates, error frequencies, and execution efficiency for two or more different use cases or business processes; comparing the performance metrics across two or more different use cases or business processes for the agent; generating one or more agent health scores comprising a metric based on comparing the performance metrics and one or more agent health score definitions that identifies an execution performance of the agent; and executing the agent with respect to the runtime based on a confirmation of the one or more agent health scores of the execution performance of the agent.
13. 13 . The hyper-automation system of claim 12 , wherein the hyper-automation system identifies the agent for the agent health score generation.
14. 14 . The hyper-automation system of claim 12 , wherein the hyper-automation system identifies an action or a task within a workflow of the one or more agentic automations assigned to the agent.
15. 15 . The hyper-automation system of claim 12 , wherein the one or more agent health scores comprise a metric measuring agent performance as the agent executes the one or more agentic automations.
16. 16 . The hyper-automation system of claim 12 , wherein the hyper-automation system generates the one or more agent health score definitions for measuring the success rates, the error frequencies, and the execution efficiency.
17. 17 . The hyper-automation system of claim 16 , wherein the one or more agent health score definitions comprise evaluation infrastructure, runtime performance, developer experience, security and compliance, and business impact.
18. 18 . The hyper-automation system of claim 12 , wherein the performance metrics align with organizational priorities that provide customizable weightings.
19. 19 . The hyper-automation system of claim 18 , wherein the customizable weightings comprise a scale from zero (0) to one hundred (100) with a scoring criteria represented as a percentage.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS This application claims priority to IN application No. 202511035107, filed Apr. 10, 2025, to IN application No. 202511035224, filed Apr. 10, 2025, and to IN application No. 202511035258, filed Apr. 10, 2025, the contents of which are hereby incorporated by reference in their entirety. FIELD The present invention generally relates to agentic automation, and more specifically, to agent health score for agentic automations. BACKGROUND Conventional automation technologies are subject to problems and errors. By way of example, problems and errors can include automation hallucinations, a lack of standardized metrics to assess automation production readiness, difficulty in comparing automation performance across diverse use cases, insufficient visibility into runtime behavior and early detection of potential issues, complex compliance and security requirements for automation decision processes, and limited capability to measure and optimize an impact of automations on business outcomes. Accordingly, an improved and/or alternative approach may be beneficial. SUMMARY Certain embodiments of the present invention may provide alternatives or solutions to the problems and needs in the art that have not yet been fully identified, appreciated, or solved by current technologies and/or provide a useful alternative thereto. For example, some embodiments of the present invention pertain to agentic automations for feedback operations with completion verification, agent accuracy evaluations, and agent health score generation. According to one or more embodiments, the method is provided. The method is performed by a hyper-automation system performing one or more agentic automations. The hyper-automation system include a memory storing for the one or more agentic automations and for agent health score generation. The hyper-automation system includes at least one processor executing the computer code to cause the method to optimize the one or more agentic automations. The method includes generating a set of standardized metrics for application to an agent executing one or more agentic automations. The method includes performing an evaluation of the agent executing the one or more agentic automations that determines performance metrics by measuring success rates, error frequencies, and execution efficiency for two or more different use cases or business processes. The method includes comparing the performance metrics across the two or more different use cases or business processes for the agent. The method includes generating one or more agent health scores based on comparing the performance metrics and one or more agent health score definitions. According to one or more embodiments, the method is provided. The method is performed by a hyper-automation system performing one or more agentic automations. The hyper-automation system including a memory storing computer code for the one or more agentic automations, for one or more evaluation sets, for agent health score generation and enhancement. The hyper-automation system comprising at least one processor executing the computer code to cause the method to refine and improve over time the one or more evaluation sets. The method includes determining whether an evaluation set of the one or more evaluation sets for an agent is acceptable. The method includes, when the evaluation set is determined as acceptable, executing the evaluation set for the agent. The method includes, when the evaluation set is determined as not acceptable, enhancing the evaluation set for the agent by sufficiency determinations or health score assessments through iterative improvements until the evaluation set is acceptable. According to one or more embodiments, the method is provided. The method is performed by a hyper-automation system performing one or more agentic automations. The hyper-automation system includes a memory storing computer code for the one or more agentic automations, for one or more evaluation sets, and for evaluation score generation and enhancement. The hyper-automation system comprising at least one processor executing the computer code to cause the method to refine and improve over time the one or more evaluation sets. The method includes determining whether an evaluation set of the one or more evaluation sets for an agent is acceptable. The method includes running an evaluation set of the one or more evaluation sets for an agent. The method includes generating an evaluation score based on the running of the evaluation set. The method includes, when the evaluation score is determined as acceptable, flagging the agent as ready for production; The method includes, when the evaluation score is determined as not acceptable, iteratively executing multi-touch attribution for assigning which factors contributed to a low evaluation score, determining an optimization for the factors, and implementing the optimizations and reruns the evaluation set until the