Unleash the Power of Build Agents in the #BuildWithBuildAgent Challenge

[laurentapia]

Our annual #BuildWith Challenge is back for the 4th year, and we can’t wait to see what you create using ServiceNow Build Agent. Build Agent is Now Assist’s agentic AI tool designed to do all the heavy lifting of building  a fully featured ServiceNow application via natural language. #BuildWithBuildAgent is your chance to vibe code with ServiceNow and see just how fast you can build an amazing app! 

Join the challenge to get recognized and inspire the ServiceNow Community with what you can build using Build Agent. 

Complete the Challenge

 

What is Build Agent? 

ServiceNow Build Agent is a smart AI tool that helps you build, edit, and deploy full-stack applications via a conversational interface. It’s designed to support developers of all backgrounds, whether you're just starting out or a veteran of the platform. 

Build Agent is what we’ve been calling an accelerating tool; it accelerates your build process so you can bring ideas to life faster. It writes the code, organizes your files, fixes errors as it goes, and takes care of both the core logic and the user interface components. 

Just describe a task using natural language, and the agent manages the entire build process. 

Why Join the #BuildWithBuildAgent Challenge? 

• Get hands-on with ServiceNow’s innovative agentic AI tool 
• Inspire others by sharing what you built 
• Earn recognition in the Community

How to Participate: 

1. Get Started: Go into your ServiceNow IDE in your instance. Once you have activated the Build Agent skill (pre-installed on Personal Developer Instances), you can create a new application or edit an existing ServiceNow IDE app using natural language prompts. Include the initial prompt that kicked off your Build Agent journey, and any standout prompts that helped bring your app to life. 
2. Document: Share your experience with others by submitting a description (300 words or more), a quick video, or an infographic about your app on social media with the hashtag #BuildWithBuildAgent. 
3. Share: Comment on this blog post and be sure to attach a link to your social post. Include a brief description of your experience using Build Agent!

For more information on how to get started, feel free to check out the FAQs at the bottom of this post or watch the video walkthrough below. 

Getting Started with Build Agent 

Need help getting started? Check out the video below to learn how to get started with ServiceNow Build Agent and how to enter the #BuildWithBuildAgent Challenge. 

https://devlink.sn/EarlBuildWithBlog 

We’ve also compiled a helpful list of resources in our FAQs to help you get started. Explore how to participate, content creation tips, and more below. You can also access our #BuildWithBuildAgent toolkit for comprehensive guidance and best practices here. 

Still have questions? Our Community is here to support you every step of the way. Feel free to ask a question on this blog or our Now Assist forum. You can also reach out to us at buildwith@servicenow.com. Please note that response times may vary! 

Get started today and show us what you can build with Build Agent! 

General FAQs

Where Do I Enter?  

To enter, please follow all official guidelines and submit your entry by replying to this Community blog post. Your response should include a description of what you created using Build Agent, along with a direct link to your social media post where you shared your creation! Make sure to include #BuildWithBuildAgent in your social media post to enter the challenge. 

How can I get started? 

Access Build Agent in Personal Development Instance on developer.servicenow.com and sign in or create a free account. Click "Request Instance," choose the latest release, and submit your request. You'll get a URL and admin credentials within minutes to start building. 

How many natural language prompts can I use? 

For this challenge, users with Now Assist for Creator licenses have no limit! For everyone else, your ServiceNow Instance will be limited to 25 prompts. PDIs on developer.servicenow.com have a limit of 10 prompts per instance. Want to keep building with Build Agent? Just back up your work, release your Personal Developer Instance, request a new one, and continue creating your application. 

Where and What Do I Post on Community?  

Post it right here! At the bottom of this blog, leave a comment with a link to your social post, along with a description of what you’ve built using Build Agent. 

Do You Have Any Tips for Content Creation? 

We have compiled resources to help guide you through the #BuildWithBuildAgent content creation process here! 

Where and What Do I Post on Social Media?  

You can post on any social media platform, whether it’s LinkedIn, YouTube, X, Instagram, or public forums such as Reddit or Stack Overflow. You’re also welcome to share what you built using Build Agent on a personal or company blog or website. Just make sure your post is publicly accessible and includes #BuildWithBuildAgent ! 

Here’s an example of a social post: 

What if I Don’t Have Social Media or Don't Want to Post Publicly?  

Not to worry! Please send an email to BuildWith@servicenow.com, and we will contact you to redeem your swag.  

When Will I Receive My Swag?  

To receive swag, you must have completed the challenge by November 23, 9:00 AM PST, and placed your swag order no later than November 24th, 9:00 AM PST. Swag was available only while supplies lasted.

You can expect your swag to arrive in late January or early February. If you’re located outside the United States, please allow additional time for shipping. 

Do I Need to Fill Out a Submission Form?  

No submission form is required! Just follow the guidelines above and you’re good to go!  

How do I get featured in the App Gallery? 

Include the initial prompt that kicked off your Build Agent journey, and any standout prompts that helped bring your app to life. 

These prompts give us insight into your creative process and will help the team behind Build Agent choose which apps to showcase in the App Gallery. 

When Does the Challenge End?  

While the challenge officially concluded on November 24th, we welcome you to continue sharing your apps past this date. 

Who is Eligible to Enter the Challenge?  

We invite everyone aged 18 and above to join us! Whether you're an absolute beginner or an experienced developer, your contribution is valued! 

I Have a Problem Who Do I Reach Out to? 

Please comment directly on this blog or email buildwith@servicenow.com with #BuildWithBuildAgent Challenge in the subject line. Please refrain from emailing your entries to this address! We are monitoring the comments for your submissions.  

Do you have an example of what I should share on this Community blog?  

Yes! Here is an example for you. Be sure to include a link to your social media post! 

Technical FAQs

What are the usage limits? 

For Personal Developer Instances (PDI), there is a limit of 10 prompts per instance per month. For customer instances that acquire Build Agent from the ServiceNow Store, there is a limit of 25 prompts per instance per month. For instances that are licensed for Now Assist for Creator, limits are based on your license. 

When are usage limits reset? 

PDI usage limits for Build Agent prompts reset every 30 days starting from the day on which the limit is set. 

What if I run out of usage or used all my prompts before this challenge? 

The limit is per instance, so you can elect to release your current PDI and request a new one, which will reset the usage count. Remember to back up any important work on your instance before you release it. 

How long can each prompt be? 

Very long. Feel free to be as simple or verbose as you want in each prompt. You can copy and paste large console logs, lengthy Product Requirement Documentation, or even huge error messages.  

What else does Build Agent handle besides writing the code? 

• Compiling and validating source files 
• Testing of its own files 
• Pushing builds into the instance 

How does Build Agent differ from AI Agents or Agentic AI? 

While both names share “Agent,” Build Agent focuses on dev automation, whereas AI Agent (Agentic AI) focuses on task automation within workflows. Build Agent is an AI Agent. 

What are the limitations of Build Agent? 

Generally speaking, Build Agent is currently limited to the following abilities: 

• ServiceNow Fluent via the ServiceNow SDK  
• React based web components for user interfaces 

To see what ServiceNow metadata that Build Agent is able to create and edit on your behalf (like business rules, client scripts, tables, etc.), check out the documentation for Fluent here:

https://www.servicenow.com/docs/bundle/zurich-application-development/page/build/servicenow-sdk/conc... 

That means Build Agent’s scope is currently explicit. If it is not part of the Fluent SDK, then it cannot build it. But that also means that the capabilities of Build Agent are constantly growing as more and more metadata types are being added to the Fluent framework. 

What that will look like as you build your applications: Build Agent will elect to write scripted code instead of building a Flow Designer action or flow. Or another example: telling Build Agent to build something that it doesn’t have access to and forcing it to move forward despite this might cause it to build it a different way or simulate that it built it. 

Always remember that Build Agent is utilizing generative AI and can hallucinate. 

Where can I learn more about Build Agent and how to acquire it for my company? 

Read more about the product of Build Agent and Enterprise Vibe Coding at https://www.servicenow.com/products/vibe-coding.html#faq, including questions regarding governance and how Build Agent relates to App Engine. 

Terms and Conditions  

No purchase is necessary. #BuildWithBuildAgent Challenge is open only to individuals at least eighteen (18) years old at the time of entry. The Challenge begins November 12, 2025, at 10:00 a.m. Pacific Daylight Time ("PDT") and ends on November 24, 2025, at 11:59 p.m. PDT.  

Entrants must respond to the official forum with a 1) Screenshot, graphic, or video showing their app built using Build Agent or 2) a 300 minimum word description of the app they built with a screenshot using Build Agent AND entrants must post their video or blog into the ServiceNow Community Developer forum here AND share it on LinkedIn or Twitter with #BuildWithBuildAgent Challenge on their post AND the social media post will be linked to their forum post.   

The Submission must be your own, original work; The Submission must contain a relevant reason/response to the prompt of Build Agent; The Submission must meet the format and size requirements of the social platform used to enter; The Submission must conform with any applicable social media platform terms.  

The prize includes the official 2025 #BuildWithBuildAgent tote bag, while supplies last. The brand of tote bag may vary. Tote bag may take up to 4 weeks or longer to arrive once ordered. Approximate Retail Value (“ARV”): $10 USD.  

The ServiceNow Community reserves the right to access email data provided by the entrant in order to reach out to individuals who have completed the challenge. Entrants who have met the guidelines and have successfully completed the challenge will receive an email from ServiceNow with instructions on how to redeem the prize. International country restrictions will apply.  

Except where prohibited, participation in the Challenge constitutes each entrant's consent to use submission’s name, and photograph (including, without limitation, entrant’s Twitter and LinkedIn profile photo), voice, opinions and/or hometown and state for promotional purposes in any media, worldwide, without further payment or consideration. 

[AnjalPDijo]

Hello Community,
This was the first time participating in this challenge and it was really an incredible experience working with Build Agent. I just can't believe I created an app within minutes.
Here's what I created:

App Name: Mission Control

Description:
Mission Control is a ServiceNow application built using Build Agent. It simulates space mission management and includes:

• A dashboard for system health, incidents, and changes.
• Navigation to spacecraft systems, telemetry data, and mission incidents.
• An AI Mission Guide for continuous assistance and preset responses.
• Analytics for system status, telemetry trends, and incident severity.

You will be able to try out these features:
💬 Chat with AI Mission Guide:

• Click the glowing chat icon (🤖) in the bottom-right corner
• Ask questions like:
  • "How are space missions going on?"
  • "What's the system status?"
  • "Show me the power levels"

🧭 Use Navigation Panel:

• Click any navigation button to open table views
• Click the analytics button (📈) for detailed charts

📊 Explore Analytics:

• Monitor risk score distribution across systems
• See detailed system metrics with real-time data

📝 Record Management:

• Click records to view/edit details
• Risk scores auto-calculated and displayed

✨ What Makes This Special:

• AI-powered responses using mission data
• Real-time dashboard updates
• Immersive space theme
• Responsive design

Screenshots:

Prompts I Used:

Create a new application called "Space Mission Control" with the following:

1. Tables:
- Spacecraft Systems: Fields → System Name (string), Status (choice: Active, Maintenance, Critical), Temperature (decimal), Power Level (decimal), Last Maintenance Date (date)
- Mission Incidents: Fields → Incident ID (auto-number), System Name (reference to Spacecraft Systems), Severity (choice: Low, Medium, High, Critical), Description (string), Status (choice: Open, In Progress, Resolved)
- Mission Changes: Fields → Change ID (auto-number), System Name (reference to Spacecraft Systems), Change Type (choice: Software Update, Hardware Replacement, Calibration), Scheduled Date (date), Status (choice: Scheduled, Completed)
- Telemetry Data: Fields → Timestamp (date-time), System Name (reference to Spacecraft Systems), Temperature (decimal), Power Level (decimal), Signal Strength (decimal)

2. Dashboard:
Create "Mission Control Dashboard" with:
- Real-time telemetry chart showing Temperature and Power Level trends.
- Widget showing open Mission Incidents grouped by Severity.
- Widget showing upcoming Mission Changes sorted by Scheduled Date.
- Summary widget showing spacecraft system health (Active vs Critical count).

3. Workflow:
Create a flow called "Critical Incident Response":
- Trigger: When Mission Incident Severity = Critical.
- Actions: Notify Mission Control team via email and in-app notification, suggest initiating recovery workflow, create related Change Request.

4. AI Guide Agent:
Add Virtual Agent topic "AI Mission Guide" with:
- Suggest maintenance if Last Maintenance Date > 30 days.
- When incident logged, suggest knowledge articles or auto-create change request.
- Quick actions: "Generate Recovery Workflow", "Show Telemetry Trends", "Predict Risk Score".

5. Mock Data:
Populate Spacecraft Systems with 5 sample systems: Propulsion, Navigation, Communication, Thermal Control, Power Supply.
Add 10 telemetry records per system with varying temperature and power levels.
Create 3 sample incidents (one Critical) and 2 scheduled changes.

6. Theme:
Apply futuristic space design:
- Dark backgrounds (deep blue, black, purple) with neon accents (cyan, magenta).
- Header with galaxy image and mission logo.
- Glowing borders and animated charts on dashboard.
- Rounded buttons with hover glow effects.
- Icons for spacecraft, satellites, planets in widgets.

7. Predictive Analytics:
Add calculated field "Risk Score" for Spacecraft Systems based on Temperature and Power Level thresholds.
Display Risk Score on dashboard and trigger alert when score > 80.

Then:
Update the Space Mission Control app with these enhancements:
- Add Virtual Agent chat icon fixed at bottom-right.
- Add preset response for "How are space missions going on?"
- Add navigation buttons for tables and analytics dashboard.
- Create UI pages for each table with record details and analytics widgets.
- Apply futuristic space theme to all new pages and charts.

Finally, I was truly impressed!
Thankful to the community and the ServiceNow team for implementing this feature. Really enjoyed working on it!
#BuildWithBuildAgent #ServiceNow

[btippens2]

🎉Process Waste Finder Application: Summary of Completion

The comprehensive Process Waste Finder application has reached 100% completion, with all core features, the manager dashboard, and analytics fully implemented. The application is reported as production-ready.

🎯Key Implemented Features

The core requirements are fully met, including:

• Custom Tables ($u\_objective\_time\_tracking$, $u\_sla\_gap\_config$) extending the task SLA.

• Hierarchical Security Roles ($sla\_gap\_analyst \to sla\_gap\_manager$) enforced by comprehensive Access Control Lists (ACLs).

• An Intelligent Business Rule for automated task tracking and SLA definitions with an 8-5 weekday schedule.

🖥️ Manager Dashboard & Visualization

A responsive React UI Page Manager Dashboard was built using a warm pastel palette as specified. It features several key reports and gauges:

• Process Waste Report: Uses a distinct color-coding system: 🔴Red for Positive Waste (Inefficiency), 🟣 Purple for Negative Waste (Agent Overwork), and 🟢 Green for Balanced Processes.

• SLA Metrics Gauge: Compares average contractual time versus true work time.

• Zombie Task Report: Identifies tasks with zero progress.

• Threshold Configuration with live editing and a "Show All Waste" toggle.

📊Data, Analytics, and Structure

The application includes a robust process waste calculation engine and automated business duration tracking. Over 20 comprehensive demo records have been created to cover all possible scenarios, including severe positive waste, overwork, and balanced examples. The complete application structure, including the menu, professional styling, CRUD support, and documentation, is finished.

***PDIs do not have Task Mining enabled. The modified approach below can still work, but intended results may vary***

Creating an "Objective Clock" without the Task Mining plugin on your Zurich PDI—requires replacing the Task Mining data source with a custom table designed to capture and measure True Work Time.

Here is a structured, prompt-efficient plan designed to build your core application functionality in 5 to 6 Build Agent prompts, ensuring you stay within the 10-prompt limit for your Personal Developer Instance (PDI).

Architectural Goal: Replacing the Task Mining Data Source

The goal of your application, based on our previous discussion, is to compare the "Contractual Clock" against an "Objective Clock" to identify Process Waste.

Clock Type	Source in PoC (Simulated)	Purpose
Contractual Clock	Existing task_sla table	Measures SLA Elapsed Time based on platform state changes (e.g., Incident state = Awaiting User Info).
Objective Clock	New Custom Table & SLA Definition	Simulates the True Work Time (hands-on-keyboard time) by tracking specific agent actions (instead of desktop recordings).

Since you cannot use Task Mining (sn_tm_core_user_data), we will use Build Agent to create a new custom table, which we can call the u_objective_time_tracking table, and configure it with its own SLA logic to serve as your alternate SLA table, representing the Objective Clock.

Build Agent Proof of Concept Plan (5-6 Prompts)

To maximize the efficiency of your prompts, you will use a robust initial prompt format that includes requirements for tables, roles, and basic logic.

Prompt 1: Initial Application Creation and Data Model (1 Prompt)

The first prompt establishes the application scope, the primary tables, security roles, and initial forms.

Action	Prompt Content & Detail	Rationale
Create App Scope & Tables	"Create a new scoped application called 'SLA Gap Analyzer.' Define two custom tables: 1) u_objective_time_tracking (extending task_sla or similar base table) with fields for u_work_start_time, u_work_end_time, u_total_true_work_duration, and a reference to the main task table. 2) A configuration table u_sla_gap_config to hold thresholds."	This establishes the custom SLA table requested for the Objective Clock and its necessary linkage to the parent task.
Define Roles	"Define two security roles: sla_gap_analyst (read access) and sla_gap_manager (read/write/report access to all tables)."	Ensures proper security architecture from the start.
Initial Forms	"Create a basic form view for the u_objective_time_tracking table, listing all defined fields."	Provides immediate visual confirmation of table structure and fields.

Prompt 2: Objective Clock Logic (1 Prompt)

This prompt configures the logic for the custom Objective Clock, defining when the "True Work Time" should accrue.

Action	Prompt Content & Detail	Rationale
Define SLA Logic	"Create a new SLA definition targeting the u_objective_time_tracking table. Set the Start condition to 'Parent Task State changes to In Progress' AND a new custom field u_agent_active becomes true. Set the Stop condition to 'Parent Task State changes to Complete/Closed' OR u_agent_active becomes false. This SLA must calculate and store the business duration on the u_total_true_work_duration field."	This simulates the Task Mining mechanism by creating a new contractual clock that only runs based on a defined agent active condition, providing the required alternative to the task_sla table.
Business Rule	"Create a business rule on the main Task table (e.g., Incident) that automatically creates and manages records in u_objective_time_tracking based on user assignment and activity (e.g., on first save/update by the assigned user, set u_agent_active to true)."	Ensures automation between the task and the new objective clock table.

Prompt 3: Calculating Process Waste (1 Prompt)

This prompt focuses on aggregating the data from the two distinct clocks to calculate the core metric: Process Waste.

Action	Prompt Content & Detail	Rationale
Data Comparison View	"Create a custom Database View (e.g., u_sla_gap_view) that joins the original task_sla table (aliased as 'Contractual') and the new u_objective_time_tracking table (aliased as 'Objective') on the task sys_id."	A Database View is an in-platform architectural solution for complex aggregation.
Calculation Field	"Add a calculated field to this view, named 'Process Waste Duration,' which calculates the difference between the Contractual's business_elapsed_time and the Objective's u_total_true_work_duration."	Directly implements the calculation: Process Waste = SLA Elapsed Time - True Work Time.

Prompt 4: Manager UI and Reporting (1 Prompt)

This prompt creates the user interface necessary to operationalize the PoC's findings, similar to the "Zombie Task Report" discussed previously.

Action	Prompt Content & Detail	Rationale
Create Dashboard/Workspace	"Build a simple UI Builder dashboard for the sla_gap_manager role, using a warm pastel palette, focusing on three key reports from the u_sla_gap_view:"	This creates the visual interface necessary to present the data.
Key Metrics	"1) A list report showing tasks where 'Process Waste Duration' exceeds the threshold in u_sla_gap_config. 2) A gauge chart showing 'Avg. SLA Elapsed Time' vs. 'Avg. True Work Time' across the last 30 days. 3) A report detailing tasks 'In Progress' with zero u_total_true_work_duration (the simulated 'Zombie Task Report')."	Delivers the critical side-by-side comparison (Metric 1 vs. Metric 2) necessary for the diagnosis phase.

Prompt 5 & 6: Refinement and Error Correction (2 Prompts Reserved)

You have two prompts remaining (or four, if you use only five) to handle debugging, documentation, or minor aesthetic adjustments.

• Prompt 5 (Refinement): Use this for specific UI tweaks (e.g., "Adjust the dashboard to use a better visualization for the Process Waste trend") or adding sample data.

• Prompt 6 (Error Correction): If Build Agent encounters an issue (e.g., a script error, or a form error), copy the error text and submit it to Build Agent: "When I load the [Specific Form], I see this error: [Paste Error]. Please fix it and deploy the change".

By using this structured approach, you ensure your PoC meets the technical requirement of comparing the contractual clock against a derived objective clock while adhering to the prompt limits of the PDI.

--------------------------------------------------------------------------------

Analogy for Understanding the Objective Clock Substitution

Think of the standard task_sla table (the Contractual Clock) as a bus schedule. It tells you when the bus should start and stop based on fixed rules (the SLA definition). If the bus is "on pause" due to a road closure (a contractual condition), that time is excluded from the schedule calculation.

Since you don't have Task Mining (the equivalent of a GPS tracker showing the bus's exact, physical movement), you are creating a driver logbook (u_objective_time_tracking table). This logbook is a custom form of tracking that starts and stops based on defined driver actions (when the agent explicitly starts working on the task). By comparing the bus schedule (SLA) to the driver's logbook (Objective Clock), you can objectively measure how much time was lost to inefficiency or clerical errors (Process Waste)

[VaranAwesomenow]

App Name: Onboard Tenants to Technical Marketplace

First Prompt: Create an app to onboard Tenants to Technical marketplace

Second Prompt: Create a service portal as entry point for Tenants

Third Prompt: Tenants use the marketplace to onboard to various service provider services such as Single SignOn, Gitlab, Cloud service account, create required order guide to create a journey for the tenant Approval is needed from each of the service provider's operations manager before tenant can be onboarded. Onboarding of tenant to service provider is automated, via flow integration to 3rd party API create a framework to handle that Tenant also has various teams such as fulfillment group, Technical approval group, business approval group and also project management group. Each tenant generally onboards a project so we need a table to store that and associate that to tenant Leverage CSDM datamodel to meet above requirements. Once tenant has system administrators who will be part of fulfillment groups who will be onboarded to service providers and will be granted roles such as administrator, developer include a way to track this information.

Human in the loop: 

The deployment requires a metadata sync before proceeding. This is common when there have been external changes to the ServiceNow instance. Would you like me to:

1. Proceed with deployment (this will sync and deploy the enhanced tenant onboarding framework)
2. Skip deployment for now and provide a summary of what's been built
3. Something else?

Fourth prompt: It did a good job initially but subsequent prompts resulted in error. I found that it doesnt do a good job with roles field and frequency field on catalog item, service catalog and catalog categories table.

After cleaning up error files I managed to get the app deployment successful.

GitRepo Link: https://github.com/anilvaranasi/TechMarketPlaceBuildChallenge.git

Screenshots of the app
Portal : 

[Chetan Mahajan]

🌟 Just Wrapped Up the #BuildWithBuildAgent Challenge – Built a Complete Employee Transport Management App in Minutes!😍

I just completed the #BuildWithBuildAgent #challenge and I’m honestly blown away
Using only a 3–4 line prompt, I was able to generate a full Employee Transport Management application — complete with beautiful UIs for Employees, Transport Admins, and Drivers.

The experience was unbelievable. Build Agent created everything in minutes:
✨ Modern responsive UI
✨ Front-end for all three personas
✨ Workflows
✨ Logic
✨ Security
✨ Even post-build diagnostics and automatic issue fixing!

---

🚖 What the App Includes

👨‍💼 Employee Portal

• Raise Pick & Drop transport requests

• Select location

• View assigned driver details

• Track driver status / dummy navigation screen

🛠 Admin Portal

• Beautiful dashboard to manage day-to-day transport operations

• Add & manage driver profiles

• View employee requests

• Assign drivers instantly

• Monitor trip workflow in real-time

🚗 Driver Portal

• View today’s pickup and drop list

• Access employee details

• Update trip status

• Dummy map to navigate to employee location

The entire application was built using HTML, CSS, JavaScript, JSX, TypeScript, and React JS — auto-generated by Build Agent. Unreal! 🙌

---

🧠 Most Impressive Part

After creating the app, Build Agent automatically ran diagnostics and fixed issues.
No need to manually debug — just Review → Approve/Reject → Done.
This truly feels like the future of development.

---

📊 What’s Working

• ✅ Backend tables deployed and accessible

• ✅ Automatic driver profile creation when driver role is assigned

• ✅ Fully functional Admin → Driver assignment flow

• ✅ Real-time status updates throughout workflow

• ✅ Beautiful modern UI with animated components

• ✅ Fully implemented ACLs and role-based access
  
  

[Animesh Das2]

Hello Community,

As part of this year's #BuildWithBuildAgent challenge, I used ServiceNow’s Build Agent to create 'Animal Rescue Management' app to help NGOs that rescues, vaccinates, rehabilitates, and rehomes stray animals. This entire app along with logics and enhanced UIs got created automatically by the Build Agent without a single line of coding. Just Amazing!

LinkedIn Post: https://www.linkedin.com/posts/animesh-das-09361bb0_buildwithbuildagent-servicenow-buildagent-activi...

Here is the step by step walkthrough of how I built the app in my PDI.

Step 1: Define the App Concept

I started with a simple goal to build an app for animal rescue NGO: Build an interactive and one stop portal for Animal Rescue management where an NGO can manage their animal rescues end to end along with having public access for rescue request submissions and online donations.

The core requirements are:

✅ Enhanced UI experience: User friendly interactive One stop portal to manage everything.
✅ Rescue Request System: geolocation, photo upload, real-time status updates.
✅ Animal Lifecycle Tracker: detailed animal profiles with medical history, progress timeline, documents, and adoption stage.
✅ Volunteer Task Assignment: assign missions with routes, instructions, and deadlines.
✅ Adoption Portal: highlight animals ready for adoption, allow adoption applications.
✅ Quick stats chart on Animals rescued: Shows combined different status of animals.
✅ Donation Management: online payments, donor tracking, receipt generation, monthly reports.
✅ Success stories: Features some success stories of animal rescues on portal.

Step 2: Prompt to the Build Agent

I fed the well structured prompt to Build Agent:

"Create 'Animal Rescue Management' scoped app from scratch for an NGO, that rescues, vaccinates, rehabilitates, and rehomes stray animals. The app will have an interactive portal UI and it should allow the NGO team having full access to track every rescued animal right from the moment it is found, brought to shelter, vaccinated, treated, rehabilitated, and ultimately adopted. Local people should have limited access to the portal be able to submit rescue requests with location, and urgency details and track their own requests.

The portal should show some success stories, analytics and charts based on animals rescued, recovered, adopted, donations, etc.

At the end create relevant application menus and modules to access the application and it's components from the left navigation."

Build Agent generated the entire application - UI pages, data model, logic, and portal configuration—based solely on this prompt.

Step 3: Debugging & Prompt refinement

While building the app the agent faced some errors and it tried multiple ways to resolve those and sometimes asked my suggestion as well in terms of natural language prompts.

I also saw some requirements were not implemented the way I was looking for so I had to refine my prompts with more clear instructions.

For example, the cards on the portal UI were not clickable and were not redirecting to the actual records and for that I used the below prompt,

"Please make the portal more interactive with making the cards clickable to redirect to the relevant records and also have 'Back' with left arrow button on top left corner of each page that open after any button or card click." 

The Build Agent analyzed the prompt and did the magic beyond my expectation.

Special Thanks to @Earl Duque @kristymerriam @laszloballa @Travis Toulson @laurentapia 

Here are the infographics:

[ravi_ramod2]

Hello Everyone,

completed the #BuildWithBuildAgent Challenge.

I just built a 'Restaurant Runner Application' using the #ServiceNowBuildAgent — created end-to-end through simple natural language prompts

The app is having a customized dashboard for restaurant owners, displaying key performance indicators (KPIs) and real-time data.

I have use the below prompt to create a whole application.

"Design a ServiceNow application named 'Restaurant Runner' with the RR Logo as the application icon. The app should have a customized dashboard for restaurant owners, displaying key performance indicators (KPIs) and real-time data. The dashboard should include:

1. *Sales Overview*: Total sales, revenue, and sales trends.
2. *Inventory Alerts*: Low stock alerts, inventory levels, and reordering suggestions.
3. *Customer Insights*: Customer satisfaction ratings, loyalty program participation, and feedback.
4. *Order Status*: Real-time order status updates, including pending, in-progress, and completed orders.
5. *Table Status*: Table occupancy, reservations, and waitlist updates.

The application should include the following features:

1. *Order Management*: Manage dine-in, takeaway, and delivery orders in one place.
2. *Billing & Invoicing*: Generate bills, process payments, and send invoices to customers.
3. *Inventory Management*: Track stock levels, automate reordering, and optimize inventory.
4. *Sales Analytics*: Provide insights on sales trends, best-selling items, and customer preferences.
5. *Customer Management*: Store customer data, track loyalty programs, and send personalized offers.
6. *Table Management*: Manage table reservations, waitlists, and seating arrangements.
7. *Reporting & Analytics*: Offer real-time reports on sales, inventory, and customer feedback.

The application should be user-friendly, accessible on mobile and desktop, and integrate with popular payment gateways and delivery services."

This updated prompt includes the RR Logo and a customized dashboard for the restaurant owner, providing a comprehensive overview of the restaurant's operations.

Click on the link below to see what agent is made by using my ideal prompt.

Link : https://youtu.be/HLg6BvodUwQ?si=Q-j27byhhMZyQ5CS

Results are very impressive : AI-assisted development made the entire process super fast and easy — what usually takes days was built in just minutes!

#servicenow #BuildAgent #prompt #Ai #chellenge #servicenowNewFeatures #keeplearning #keepGrowing

[anandsuresh]

Introduction:

Life is full of chaos, and what if I tell you that handling the major incidents in the big companies is equally catastrophic? 😂

If you are working in a support project, you would definitely know how crucial every minute is when your production application goes down. Also, if you are part of the Major Incident Management (MIM) team, then you would know exactly how stressful it is, and sometimes feel helpless. I built this to make major incidents feel less like a horror movie 👹 and more like a well-planned rescue mission. Faster response. Smarter decisions. Fewer heart attacks.

What is MICC?

I have built an application as part of the #BuildWithBuildAgent challenge, but it turned out to be surprisingly good. 

The name of the application is Major Incident Command Center, and as you have guessed correctly by the name, this is a one-stop place for the MIM team to have a complete overview of the major incidents to seamlessly handle them without panic attacks

Core Features:

When we build an application, we need to first understand what problem we are going to solve. Right now, the MIM team is almost feeling helpless due to scattered information and has to do everything manually. So, I have added the following features by keeping in mind the pain points faced by the MIM and support teams.

✅ Complete overview of the dashboard

✅ Predict the Major Incidents proactively

✅ War Room for centralized collaboration

✅ Business Communication and active updates

✅ On-call information and escalation Matrix

✅ Admin Management to configure metrics

Problem and Solution:

You can find the screenshot of the Major Incident Command Center application below with all the above-mentioned menus.

The main problem faced by the business is the delayed restoration of the impacted services, and the MIM team faces delays due to late reporting, wrong prioritization.

So, I am bringing a solution to many problems in one place, and you can not only manage the end-to-end major incidents in one single place, but you can also predict the major incidents before anyone even knows. 

Now, you can alert the support teams through various channels as below.

• SMS Alert
• Email Alert
• War Room Notification
• Teams Alert

Video Demo:

Conclusion:

So that is my Major Incident Command Center. 😊

A single place to stop the chaos, rescue the business, and let the MIM team breathe for once. 🧘

I am proud of how far this has come using only natural language and a little bit of stubbornness.

And this is just the beginning. More features. More automation. More peace for the support heroes 🦸

#Servicenow #BuildWithBuildAgent

[AbinC]

Thrilled to share my #BuildWithBuildAgent project 🚀
I’ve built a complete Employee Promotion Management System in ServiceNow HRSD that streamlines the full promotion lifecycle — from nomination to payroll integration.

🏗️ Core Architecture with 3 interconnected tables
🔐Role-based security with 6 distinct roles
🤖Automated workflows for eligibility, salary updates, and evaluations
🎨React-based dashboards for HR, managers, and employees

✅Transparent nomination & approval chain
✅Automated communication & integrations with payroll/LMS

✅Tailored dashboards with actionable insights

This solution empowers HR teams to manage promotions efficiently while ensuring fairness and compliance.

👉Explore the dashboard here:  https://dev188221.service-now.com/x_1682344_employee_promotion_dashboard.do

you can find my LinkedIn post here : https://www.linkedin.com/posts/abinpk_buildwithbuildagent-servicenow-servicenowcommunity-activity-73...

hashtag
hashtag#ServiceNow hashtag
hashtag#ServiceNowCommunity hashtag
hashtag#BuildWithBuildAgent hashtag
hashtag#ServiceNow hashtag
hashtag#AppDevelopment Earl Duque Laszlo Balla Kristy Merriam Travis Toulson Lauren McManamon Sarah Garey Celina Zamora Allen Andreas Mark Obee Derek Roth Gordon Lauren Tapia Dr. Atul Grover 🧭 Learn N Grow Together 📚The ServiceNow StorytellerDinesh Kumar RaghuSandeep Dutta Community

[mitmitesh75]

I’m thrilled to share that i have successfully created a Hospital Management System Tracker using #BuildWithBuildAgent 🚀

This solution is designed to:

• 📋 Track both new and Old patient records.
• 🩺 Maintain detailed treatment histories linked to specific diseases.
• 👨⚕️ Map doctors to treatments and track their duty schedules
• 🔍 Suggest the right doctor based on specialization and disease type
• 📊 Provide dashboards for hospital administrators to monitor patient trends and doctor workloads

By combining structured data design with intelligent automation, this tracker empowers hospitals to deliver more efficient, patient-centered care while ensuring transparency and accountability.

Linkdinpost :- https://www.linkedin.com/pulse/just-complete-buildwithbuildagent-challenge-mitesh-kumar-ln03c/

This Challenge showed me how AI and low-code can democratize application development.

Huge Thanks to serviceNow for this incredible learning experience!

#BuildWithBuildAgent

#serviceNow

#ServiceNowCommunity

#LowCode

[TeoG31032040407]

Hey everyone,

Completed the #BuildWithBuildAgent Challenge. 

My approach was a bit different since I tried to build a phone app instead of a web workspace interface. I ended up creating a scope that tracks assets by scanning barcodes solely using "prompt engineering".  I used ServiceNow Studio, SN IDE, Now Mobile Studio and help myself with the Android SDK simulator and IOS simulator to test the app in both OS. I built the following template to cover the app architecture from ACL, tables, ATF, BRs to Mobile Records, Applet, Screens.  

On my first template, my focus was to build the back implementation in order to have a clear back-end service, covering the following modules:  

• Tables + dictionary fields

• Business rules / scripts

• Flows / workflow logic

• Sample/demo data

• Roles + ACL intent

• ATF scaffolding (server-side test shells)

On my second template within the same conversation, my focus was in the Mobile Experience side. Prompted Now Assist to cover the following modules:

• Mobile Experience record

• Applet

• Navigation tabs

• Screens (Scan Home, Pending Scans, History)

• Data Items (pointing to your tables)

• Actions / Functions (Scan Barcode → create Scan Event)

The following screenshot represents all the important modules that were created using Now Assist Agent in the first template. 

The following screenshot represents the mobile views for an ios device where I'm opening the camera function in order to start scanning the barcodes. The following components were created using the second template. (For the record, no android screenshot available since I forgot my Windows laptop at work)

Something I noticed is that Now Assist did not fully integrate the wiring between the Now Mobile experience and the back-end modules previously created. The following are the services that were not fully mapped by Now Assist:

1.  Offline enablement wiring

• Mark my Scan Event table as offline-capable

• Set offline max queue size

• Define auto-clear after sync

2. Action to table wiring in my Scan Event table

• Populate my required fields correctly (barcode, scan_type, timestamp, driver, route, sync_status).

• Pending to test the offline sync for the sync_status=pendingcolumn and set a history table.

The reason why I decided to build an offline asset tracking app was due to the demand at my previous workplace, having such scope would have been a blessing. Back at that time, I didn't have the enough knowledge to prompt my AI service and do the mapping layer understanding the fulfiller and requester's side, so that's something I learned throughout the time.

My social post is here: https://www.linkedin.com/posts/eberteo_buildwithbuildagent-servicenow-nowplatform-activity-739827658...

--------------------------

First Template

--------------------------

## Goal
- Allow drivers to scan asset barcodes **offline** and sync to the cloud once connectivity returns.
- Maintain accurate chain-of-custody across all boxes.

### Success Criteria
- [ ] Drivers can record scans while offline (0% dependency on connectivity)
- [ ] Sync latency < [X minutes] after connection returns
- [ ] Scan failure rate < [Y%] across all users
- [ ] Duplicate scan prevention logic validated
- [ ] All scans linked to routes, users, assets

### Non-Goals
- No manual paper auditing.
- Not scanning random assets not in buyer inventory.

---

## Audience (Roles)
- Drivers (offline scan workers)
  - Skill level: Novice
  - Usage: Daily, Mobile ONLY
- Supervisors / Fulfillers
  - Skill level: Intermediate
  - Usage: Workspace, desktop or tablet

---

## Mobile App Identity
- App Name: [Iron Mountain Asset Scan]
- Scope: [x_im_scan]
- Data sensitivity: Restricted
- Brand: Field-operations theme, large-action UI

---

## Core Capabilities
1. Scan asset barcodes **offline**.
2. Store scanned events **locally on device**.
3. Sync queued scans **to cloud** once online.
4. Validate barcodes against **authorized assets list** (route-based subset).
5. Open fulfillers’ Workspace page to manage exceptions/events.
6. Capture GPS location (optional) at scan time.
7. Display asset status visually (picked up, delivered, verified).

---

## Mobile Offline Configuration
**Offline Tables**
- [x_im_scan_event]
  - Stores barcodes + timestamps + GPS

**Device Storage Rules**
- Max queue: [1000 scans]
- Retention: Clear after successful sync

**Download Strategy**
- Preload:
  - Allowed assets for route: [Yes]
  - Locations: [Optional]
- Download method:
  - **Route-based filtering** (performance protection)

---

## Data Model

### Table: Scan Event
- Label: Scan Event
- Name: x_im_scan_event
- Fields:
  - u_barcode: string(50), required
  - u_scan_type: choice [pickup, delivery, verify]
  - u_timestamp: datetime, default now (device local)
  - u_latitude: decimal(10,6), optional
  - u_longitude: decimal(10,6), optional
  - u_sync_status: choice [pending, success, failed]
  - u_route: reference to Route table
  - u_driver: reference to User

### Table: Asset (if not native)
- Barcode unique index enforced

---

## Automation (Flow Designer)
- On sync → validate barcode exists and belongs to assigned route
- On success → update Asset state
- On conflict → create “Exception Case” assigned to Supervisor

---

## Fulfiller Experience (Workspace)
- Workspace name: [Asset Ops Workspace]
- Views:
  - Queued Exceptions
  - Verified Scans
  - Route Progress Map (optional)
- Map component:
  - Visualizes scan locations

---

## Notifications
- Sync failure alerts to driver
- Exception alerts to supervisor

---

## SLAs / KPIs
- Exceptions must be reviewed within [X hours]
- Delivery confirmation within [Y hours]

---

## Security
- Drivers: mobile-only; only see assigned routes/assets
- Supervisors: full read/write on assigned region

---

## ATF Requirements
- Validate offline → online sync chain
- Validate duplicate scan detection
- Validate exception creation logic
- Validate GPS capture

---

## Constraints
- No dependency on real-time network
- No prompts or pop-ups requiring connectivity
- If choice must be made → choose **performance over UI decoration**

--------------------------

Second Template

--------------------------

Build the missing ServiceNow Mobile Experience for my scoped app "Iron Mountai" (scope: x_1126062_iron_mou).

Host app: NOW MOBILE (not Mobile Agent).

Create a Mobile Experience + Applet with navigation and screens for DRIVERS to scan assets offline.

Use these tables:
- Scan Events: x_1126062_iron_mou_scan_event (offline-enabled)
- Assets: x_1126062_iron_mou_asset
- Routes: x_1126062_iron_mou_route
- Exception Cases: x_1126062_iron_mou_exception_case

Mobile UI requirements:
1. Create a "Scan Home" screen with a large primary action button "Scan Barcode".
   - Action uses device barcode scanner.
   - On scan, create a Scan Event locally with:
     u_barcode, u_scan_type default='pickup', u_timestamp=now (device),
     u_driver=current user, u_route=assigned route,
     u_sync_status='pending'.
   - Capture GPS if available into u_latitude/u_longitude.

2. Create a "Pending Scans" list screen:
   - Shows Scan Events where u_sync_status='pending'
   - Works offline and shows count badge on nav.

3. Create a "Scan History" list screen:
   - Shows last 7 days of Scan Events
   - Filter by current driver.

4. Add Offline configuration:
   - Enable offline for Scan Event table.
   - Download subset of Assets for driver’s assigned route only.
   - Max offline queue 1000 events.
   - Auto-clear local Scan Events after successful sync.

5. Navigation:
   - Tabs: Scan Home, Pending Scans, Scan History.

6. Push notifications (optional):
   - Notify driver on sync failure.
   - Notify supervisor group on exception case creation.

Output:
- List all Mobile App Builder records created (mobile app config, applet, screens, data items, actions, nav).
- Document any assumptions.

#buildwithbuildagent #servicenow