Submission type
Coded Agent with UiPath SDK
Name
Purabi Sarma
Industry category in which use case would best fit in (Select up to 2 industries)
Information technology and services
Complexity level
Intermediate
Summary (abstract)
RPA Sentinel Agent bridges the gap between business users and UiPath Orchestrator by enabling natural language interaction for managing RPA bots.
Detailed problem statement
In many organizations, Robotic Process Automation (RPA) initiatives have scaled rapidly across departments, resulting in a large number of bots running on UiPath Orchestrator. However, managing and monitoring these bots often remains a technical and centralized task. Business users—such as operations managers, analysts, or process owners—frequently need to check bot statuses, request executions, or stop automations due to process changes or exceptions.
Currently, these users rely heavily on automation developers or the Center of Excellence (CoE) team to perform simple actions such as:
• Checking whether a specific bot has run successfully today.
• Starting or stopping a bot on demand.
• Understanding who owns a bot or where it is located in Orchestrator.
This dependency leads to communication delays, slower decision-making, and inefficient use of developer time for operational tasks. Additionally, non-technical users often lack direct access to UiPath Orchestrator or find it difficult to navigate its interface and identify the correct process or robot name.
The absence of a natural language interface between business users and the automation environment limits accessibility, scalability, and real-time responsiveness of automation operations. There is a need for a specialist agent that can understand business language, interpret intent, map it to actual automation assets, and trigger the appropriate action automatically and securely through UiPath.
By introducing an intelligent intermediary that connects human language to RPA actions, organizations can make automation more human-centric, responsive, and accessible, empowering every business user to manage bots as easily as chatting with a digital assistant.
Detailed solution
The solution introduces an AI-driven Specialist Agent that bridges the gap between business users and UiPath Orchestrator by enabling natural language interaction for managing RPA bots. The agent allows users to query, start, or stop any bot simply by typing a request, without requiring access to Orchestrator or technical expertise.
-
Conversational Intelligence Layer (Python + LangChain)
• The agent is built in Python using LangChain for natural language understanding, intent classification, and entity extraction.
• It accepts free-form business queries such as:
o “What is the status of the bot InvoiceProcessor today?”
o “Start the bot that handles vendor payments.”
o “Stop the sales report bot.”
• The model identifies:
o Intent → (status check / start / stop)
o Bot entity → extracted from user input
o Supporting context → such as time, owner, or business process -
Context Data Layer (Context Text Raw File)
• The agent references a context text raw file, which acts as a lightweight, dynamic knowledge base containing metadata for all deployed bots.
• This file includes structured details such as:
o Bot display name and corresponding Orchestrator process name
o Folder / environment location in UiPath Orchestrator
o Developer name and email ID
o Business description or purpose of each bot
• The agent uses semantic search (via LangChain’s retrieval tools) to match the user’s query against this file and retrieve the most relevant bot record.
• This design allows the system to easily update bot information without retraining or code changes — simply by refreshing the context file. -
UiPath Integration Layer (UiPath Python SDK + RPA Bot)
• After resolving the correct bot information and intent, the agent communicates with UiPath through the UiPath Python SDK.
• It passes structured data — including bot name, folder path, developer details, and intent — to a UiPath RPA bot.
• The UiPath RPA bot performs the operational tasks in Orchestrator:
o Status Check: Retrieves job state, last run time, and result.
o Start Bot: Triggers execution in the corresponding folder.
o Stop Bot: Stops the active job for that process. -
End-to-End Workflow
-
Business user sends a natural language command.
-
LangChain agent interprets intent and key entities.
-
Agent searches the context text file to map user-friendly terms to actual bot metadata.
-
Structured details are sent to UiPath via the Python SDK.
-
RPA bot executes the operation in Orchestrator.
-
Result (bot status, start confirmation, etc.) is sent back to the user conversationally.
Narrated video link (sample: https://bit.ly/4pvuNEL)
Expected impact of this automation
• This automation makes RPA operations more accessible, responsive, and human-centric by allowing business users to manage UiPath bots through simple natural language commands. Instead of depending on developers or manually logging into Orchestrator, users can now ask the Specialist Agent questions like “What’s the status of the Invoice Bot today?” or “Start the Payroll Bot” — and receive instant results.
• From a business perspective, this reduces the communication gap between business teams and automation developers. Operational managers can take real-time action without delays, improving process continuity and reducing downtime. Routine queries and actions that previously required manual intervention are now handled automatically, saving hours of coordination time each week.
• From a technical and project standpoint, the use of a context text file enables easy scalability — new bots can be added or modified by simply updating the file, without retraining or redeploying the agent. This ensures faster onboarding of new automations and minimal maintenance.
• Overall, the Specialist Agent improves productivity, enhances transparency, and supports self-service automation. It transforms how teams interact with UiPath by blending conversational AI and RPA, resulting in smarter operations, quicker responses, and a stronger return on automation investment.
UiPath products used (select up to 4 items)
UiPath Coded Agents
UiPath Maestro
UiPath Orchestrator
UiPath Robots
UiPath Studio
Integration with external technologies
None
Hi there, 
Nov 5–16: Jury evaluation by 
Nov 17: Winners announced 
.
,