Unified artificial intelligence agent, robotic process automation robot, and agentic orchestration process development applications
Unified artificial intelligence (AI) agent, robotic process automation (RPA) robot, and agentic orchestration process (AOP) development applications are disclosed. In such embodiments, AI agents “coexist” in tandem with RPA robots that execute RPA automations. Automation developers can provide a natural language description of a process intended to be implemented by an AI agent, as well as prompts for a user. Developers can also select grounding context, service integration tools, RPA robots, and escalation attributes for the AI agent. When executed, the AI agent will implement a dynamic flow and attempt to perform its logic autonomously. If this is not successful, the AI agent can escalate to a human-in-the-loop, and the human resolution may be saved in memory for self-healing purposes. Developers can also develop AOPs that can utilize AI agents and RPA robots via a conductor application as well.
1. One or more non-transitory computer-readable media storing one or more computer programs, the one or more computer programs configured to cause at least one processor to:
display an artificial intelligence (AI) agent development interface comprising a system prompt, a user prompt, selectable tools that can be employed by the AI agent, context grounding, escalation options, and a field to run queries;
generate an AI agent based on the system prompt, the context grounding, and one or more tools selected from the one or more selectable tools, the AI agent configured to call a generative AI model, provide input to the generative AI model, and receive output from the generative AI model to process queries from the query field; and
execute an agentic automation via the generated AI agent by:
providing a query entered into the query field to the generative AI model as input and receiving results from the generative AI model as output, and
displaying the results from the generative AI model, executing the one or more tools via the generated AI agent, and displaying results of the execution of the one or more tools.
2. The one or more non-transitory computer-readable media of claim 1 , wherein the one or more selectable tools comprise one or more integration service connectors for an external application.
3. The one or more non-transitory computer-readable media of claim 1 , wherein the one or more selectable tools comprise one or more automations executed by one or more robotic process automation (RPA) robots.
4. The one or more non-transitory computer-readable media of claim 1 , wherein the one or more computer programs are further configured to cause the at least one processor to:
display an escalation configuration interface that provides human-in-the-loop configuration options; and
configure the AI agent to utilize the human-in-the-loop configuration options responsive to the AI agent not being able to successfully complete a task requested in the query field.
5. The one or more non-transitory computer-readable media of claim 1 , wherein responsive to the AI agent not being able to successfully complete a task requested in the query field, the one or more computer programs are further configured to cause the at least one processor to:
display an agent handler interface comprising a query from the AI agent for a human-in-the-loop to provide a solution for; and
store a response from the human-in-the-loop in agentic memory of the AI agent, thereby allowing the AI agent to accomplish the task in the future, wherein
the AI agent is configured to search the agentic memory when executing the agentic automation.
6. The one or more non-transitory computer-readable media of claim 1 , wherein the one or more computer programs are further configured to cause the at least one processor to:
display a trace interface from the execution of the agentic automation, the trace interface comprising an execution actions pane and an attributes pane displaying attributes of a selected execution action in the execution actions pane.
7. The one or more non-transitory computer-readable media of claim 6 , wherein the trace interface facilitates debugging of an agentic loop, inputs provided to the one or more tools, and outputs returned from the one or more tools.
8. The one or more non-transitory computer-readable media of claim 1 , wherein the one or more computer programs are further configured to cause the at least one processor to:
display an application orchestration process (AOP) development interface comprising AOP workflow components, selectable AI agents, and selectable robotic process automation (RPA) robots; and
generate an AOP workflow comprising one or more steps that are executed by one or more RPA robots and/or one or more AI agents via a conductor application.
9. The one or more non-transitory computer-readable media of claim 8 , wherein the AOP development interface comprises a field where a user can enter a description of a business process and the one or more computer programs are further configured to cause the at least one processor to:
automatically generate an AOP workflow from the business description process field using the generative AI model or another generative AI model.
10. A computer-implemented method, comprising:
generating an artificial intelligence (AI) agent, by a computing system, based on a system prompt, context grounding, and one or more tools, the AI agent configured to call a generative AI model, provide input to the generative AI model, and receive output from the generative AI model to process queries from a query field; and
executing an agentic automation, by the computing system, via the generated AI agent by:
providing a query entered into a query field to the generative AI model as input and receiving results from the generative AI model as output, and
displaying the results from the generative AI model, executing the one or more tools via the generated AI agent, and displaying results of the execution of the one or more tools.
11. The computer-implemented method of claim 10 , wherein the one or more selectable tools comprise one or more integration service connectors for an external application.
12. The computer-implemented method of claim 10 , wherein the one or more selectable tools comprise one or more automations executed by one or more robotic process automation (RPA) robots.
13. The computer-implemented method of claim 10 , further comprising:
displaying, by the computing system, an escalation configuration interface that provides human-in-the-loop configuration options; and
configuring the AI agent, by the computing system, to utilize the human-in-the-loop configuration options responsive to the AI agent not being able to successfully complete a task requested in the query field.
14. The computer-implemented method of claim 10 , wherein responsive to the AI agent not being able to successfully complete a task requested in the query field, the method further comprises:
displaying, by the computing system, an agent handler interface comprising a query from the AI agent for a human-in-the-loop to provide a solution for; and
storing a response from the human-in-the-loop in agentic memory of the AI agent, by the computing system, thereby allowing the AI agent to accomplish the task in the future, wherein
the AI agent is configured to search the agentic memory when executing the agentic automation.
15. The computer-implemented method of claim 10 , further comprising:
displaying, by the computing system, a trace interface from the execution of the agentic automation, wherein
the trace interface comprises an execution actions pane and an attributes pane displaying attributes of a selected execution action in the execution actions pane, and
the trace interface facilitates debugging of an agentic loop, inputs provided to the one or more tools, and outputs returned from the one or more tools.
16. The computer-implemented method of claim 10 , further comprising:
displaying, by the computing system, an application orchestration process (AOP) development interface comprising AOP workflow components, selectable AI agents, and selectable robotic process automation (RPA) robots; and
generating, by the computing system, an AOP workflow comprising one or more steps that are executed by one or more RPA robots and/or one or more AI agents via a conductor application.
17. The computer-implemented method of claim 10 , wherein the AOP development interface comprises a field where a user can enter a description of a business process and the method further comprises:
automatically generating, by the computing system, an AOP workflow from the business description process field using the generative AI model or another generative AI model.
18. One or more computing systems, comprising:
memory storing computer program instructions; and
at least one processor configured to execute the computer program instructions, wherein the computer program instructions are configured to cause the at least one processor to:
display an artificial intelligence (AI) agent development interface comprising a system prompt, a user prompt, selectable tools that can be employed by the AI agent, context grounding, escalation options, and a field to run queries,
generate an AI agent based on the system prompt, the context grounding, and one or more tools selected from the one or more selectable tools, the AI agent configured to call a generative AI model, provide input to the generative AI model, and receive output from the generative AI model to process queries from the query field, and
execute an agentic automation via the generated AI agent by:
providing a query entered into the query field to the generative AI model as input and receiving results from the generative AI model as output, and
displaying the results from the generative AI model, executing the one or more tools via the generated AI agent, and displaying results of the execution of the one or more tools, wherein
the one or more selectable tools comprise one or more automations executed by one or more robotic process automation (RPA) robots.
19. The one or more computing systems of claim 18 , wherein the computer program instructions are further configured to cause the at least one processor to:
display an escalation configuration interface that provides human-in-the-loop configuration options; and
configure the AI agent to utilize the human-in-the-loop configuration options responsive to the AI agent not being able to successfully complete a task requested in the query field.
20. The one or more computing systems of claim 18 , wherein responsive to the AI agent not being able to successfully complete a task requested in the query field, the computer program instructions are further configured to cause the at least one processor to:
display an agent handler interface comprising a query from the AI agent for a human-in-the-loop to provide a solution for; and
store a response from the human-in-the-loop in agentic memory of the AI agent, thereby allowing the AI agent to accomplish the task in the future, wherein
the AI agent is configured to search the agentic memory when executing the agentic automation.
21. The one or more computing systems of claim 18 , wherein the computer program instructions are further configured to cause the at least one processor to:
display a trace interface from the execution of the agentic automation, wherein
the trace interface comprises an execution actions pane and an attributes pane displaying attributes of a selected execution action in the execution actions pane, and
the trace interface facilitates debugging of an agentic loop, inputs provided to the one or more tools, and outputs returned from the one or more tools.
22. The one or more computing systems of claim 18 , wherein the computer program instructions are further configured to cause the at least one processor to:
display an application orchestration process (AOP) development interface comprising AOP workflow components, selectable AI agents, and selectable robotic process automation (RPA) robots; and
generate an AOP workflow comprising one or more steps that are executed by one or more RPA robots and/or one or more AI agents via a conductor application.
23. The one or more computing systems of claim 22 , wherein the AOP development interface comprises a field where a user can enter a description of a business process and the computer program instructions are further configured to cause the at least one processor to:
automatically generate an AOP workflow from the business description process field using the generative AI model or another generative AI model.