AgentHack submission type
Agentic Testing Solutions
Name
Vinit Kawle
Team name
BotBuilders
Team members
@Rohan_Kulkarni1 @Darshan_Sable
How many agents do you use
Multiple agents
Industry category in which use case would best fit in (Select up to 2 industries)
Information technology and services
Complexity level
Intermediate
Summary (abstract)
The current QA process for web applications heavily relies on manual efforts, where teams analyze DOM elements, document user interactions, and define validations for each functionality. Although basic automation tools like record-and-playback and rule-based bots are employed, the process remains time-consuming, error-prone, and difficult to scale. Moreover, traditional automation lacks the intelligence to adapt to frequent UI changes or understand application logic, leading to inefficiencies in test case creation and maintenance.
Detailed problem statement
In the current system, QA teams manually analyze web applications to identify test scenarios and write corresponding test cases for functionalities such as login, logout, form submissions, and data validations. This process involves:
Manually inspecting the DOM elements (IDs, classes, attributes).
Documenting step-by-step actions like entering text, clicking buttons, and verifying redirects or UI states.
Creating test data, expected results, and mapping validations.
Repeating the same process for every application change or new feature.
Basic automation might include:
Record-and-play tools to capture user actions.
Rule-based bots to populate UI with test data.
Challenges:
Time-Consuming: Creating test cases for every change consumes a large amount of time.
Human Error: High chances of missing edge cases or misdocumenting steps.
Scalability Issues: Difficult to scale for large or frequently changing applications.
Lack of Intelligence: Traditional automation cannot understand application logic or adapt to UI changes intelligently.
Detailed solution
In the current QA process, teams manually inspect web applications to identify test scenarios and create corresponding test cases for functionalities like login, logout, form submissions, and data validation. This involves analyzing DOM elements, documenting user actions, generating test data, and mapping validations repeated for each application change or new feature.
While basic automation (e.g., record-and-play tools or rule-based bots) assists with repetitive actions, it falls short in adapting to changes or understanding context.
Key Challenges:
Time-Consuming: Manual test case creation for every change is labor-intensive.
Prone to Human Error: Important edge cases or steps can be overlooked.
Scalability Limitations: Manual efforts don’t scale well for complex or frequently updated applications.
Lack of Intelligence: Traditional automation tools lack adaptability and do not comprehend business logic or UI dynamics.
This highlights a pressing need for more intelligent, scalable, and adaptive testing solutions.
Demo Video
Expected impact of this automation
The anticipated impact of this agentic automation-based solution includes:
Time Efficiency:
Reduces test case creation time by up to 80%.
Quick turnaround for regression testing.
Improved Accuracy:
AI agents reduce human error and ensure better coverage.
Scalability:
Easily extend the agent to cover other flows like registration, profile update, etc.
Adaptability:
Agents intelligently adjust to UI changes, avoiding brittle scripts.
Empowers QA Teams:
Focus shifts from repetitive work to strategic test design and risk-based testing.
Faster Releases:
Enables shift-left testing, allowing developers to receive feedback sooner in the development lifecycle.
UiPath products used (select up to 4 items)
UiPath Agent Builder
Integration with external technologies
Studio Web
Agentic solution architecture (file size up to 4 MB)
Sample inputs and outputs for solution execution
Other resources