🤖 Dream AI Agent Task – $150 prize | 24 hours onl

[gbadebo-bello]

The weekly community challenge is live!

Prompt:
If you could hand one task over to an AI agent, what would it be?

We’re talking anything from automating a mind-numbing chore to streamlining a complex workflow at work - or something totally unexpected that only you would think of. The more creative, personal, or downright clever, the better your chances of winning.

Prize:
One winner gets a $150 Postman swag pack.

BONUS
Want an extra shot at winning? Show your idea in a Postman Notebook - it can be super simple.
Just outline the steps your AI agent would take, and show a simple example of the APIs/tools it might use in a Notebook.
That’s it. Even a rough sketch works! Share the Notebook link in your reply and enter into the Notebook Challenge. Yes, you can win both.

How to enter:

• Reply to this thread with your idea during the challenge window
• Keep it clear so we can picture how your AI agent would work

Deadline:
10am BST Thursday

Winner announced Friday. Let’s hear what you’d hand off to your dream AI agent. Go!

[abdullahlko (Abdullah Ansari)]

The AI API Change Bodyguard – My Always-On API Protector

As someone who’s been jolted awake at 2 AM because a payment API suddenly changed its response, I know the chaos that follows — frantic debugging, vague error messages, and customers asking why things are broken.

That’s why my dream AI agent would be an API Bodyguard — a 24/7 smart monitor that watches every API I rely on, instantly detects breaking changes, alerts me in plain language, and even suggests quick fixes before anything goes down.

How it works:

• 24/7 Monitoring: Uses Postman Monitors or webhooks to keep an eye on all API endpoints, constantly checking for anomalies.

• Instant Impact Analysis: Spots structural changes (e.g., removed fields, altered formats) and predicts which workflows will fail — before they actually do.

• Smart Alerts: Sends crystal-clear notifications like:

  “The /orders endpoint from PaymentAPI removed delivery_estimate — this will break your order tracker in 3 hours.”

• Auto-Fix Suggestions: Generates quick patches for Postman collections or scripts to keep services running until a permanent fix is deployed.

• Calm Under Pressure: In a live incident, it summarizes exactly what changed, where, and why — so I can skip the panic and go straight to the solution.

In short, my AI API Change Bodyguard would give every developer what they secretly want: peace of mind that nothing will break without them knowing first — and a trusted partner who’s always on watch.

Postman Notebook: I made a simple Postman Notebook to show my AI agent idea in action.

Link: Notebook

[science-meteorologi2 (Kshitij Saksena)]

The Personal Assistant

AI agents excel at pattern recognition, complex reasoning, and solving problems at scale. However, they face a fundamental limitation: they only understand the general data they were trained on, not the specific details about you and your unique context.

You might assume that leveraging AI as a personal assistant requires manually inputting extensive information about your preferences, schedule, and work patterns. But Model Context Protocol (MCP) servers eliminate this barrier by creating seamless connections between AI agents and your existing productivity tools.

How It Works

Getting started with AI-powered personal assistance through Postman is straightforward:

1. Create a new AI Request in Postman: Select your preferred AI model and add its authentication token.

2. Select tools and add MCP servers: Connect your productivity tools like Google Calendar, Notion, and other platforms you already use.

3. Generate missing MCP servers: If your productivity tool doesn’t have an existing MCP server, Postman’s MCP Generator can create one using the tool’s public APIs—no coding required!

4. Start prompting: Your AI agent can now gather information from all connected productivity tools through the attached MCP servers, providing contextually aware responses.

Real-World Example

Imagine you use Google Calendar to schedule tasks and events, while maintaining Notion pages for goal tracking, skill development, and wellness monitoring. With MCP servers connected, your AI agent gains comprehensive context about your goals and schedule.

The agent can now:

• Optimize your calendar based on your stated goals

• Suggest skill-building activities during free time slots

• Recommend wellness breaks when your schedule is packed

• Provide intelligent insights by correlating data across all your tools

The Result

Instead of juggling multiple apps and manually connecting information, you get an AI assistant that truly understands your workflow, priorities, and personal context—transforming generic AI capabilities into a personalized productivity partner!

[maintenance-astrono8 (Uday)]

Event-Safety Autopilot — My Dream AI Agent

Task I’d assign to my AI agent:
I envision a real-time safety co-pilot for public events — a tireless, intelligent sentinel that continuously monitors weather conditions, crowd dynamics, and incident reports, then intelligently prioritizes and dispatches the right response before issues escalate.

Why it matters to me:
In event operations, I’ve experienced how quickly information overload can lead to missed signals. Juggling multiple data sources — weather feeds, social media chatter, emergency calls — often causes alert fatigue and delays in decision-making. My goal is to have a “calm brain” in the background: one that never sleeps, never panics, and always acts decisively.

How the Agent Works
Sense:

• Ingests real-time weather alerts via OpenWeather APIs
• Monitors live incident feeds from city/open data sources
• Receives structured on-site incident reports via Postman Mock endpoints
• Optionally processes crowd density from camera counters

Understand:

• Applies LLM-powered severity scoring and duplicate detection
• Severity policy: P0 = life-safety threat, P1 = operational/service outage, P2 = informational update

Act:

• P0: Trigger Twilio SMS/Call, log incident in Notion or Jira, and map the precise location
• P1: Send Slack notifications and log to Google Sheets
• P2: Consolidate updates into a digest every 15 minutes

Learn:

• Incorporates post-event review notes to refine thresholds and decision-making rules over time

APIs & Tools:
Postman Flows / AI Agent Builder, OpenWeather API, Twilio, Slack, Notion or Jira, Google Sheets (via Apispreadsheets or Sheety), Postman Echo/Mocks for simulation.

Bonus: I have developed a Postman Notebook demonstration that simulates the complete workflow — from incident detection to automated response.
Notebook link: Postman

[lunar-module-geolog5 (JOYSTON MONTEIRO)]

Inbox Zero - Your Personal Email Overhaul Agent

Imagine this: You open your inbox on a Monday morning… and instead of 347 unread messages, you see a clean, organized list of only the things that actually matter ,each one summarized, prioritized, and ready for action. No clutter. No spam. No guilt.

The Problem

Between work, personal, and side-project inboxes, I spend hours every week digging through irrelevant updates, promotional blasts, and “reply all” threads. Critical messages get buried. Deadlines slip. My email has become a source of stress instead of a productivity tool.

The Solution : Inbox Zero Agent

A smart AI-powered workflow, built with Postman’s plug-and-play agent templates, that turns chaotic inboxes into focused action lists.

How It Works

1. Connect & Scan

   • Securely link Gmail/Outlook/IMAP via OAuth.
   • Pull in new emails on a set schedule.

2. Understand & Prioritize

   • Apply AI-driven sentiment + urgency analysis to label each email:

     • Urgent
     • Respond Today
     • FYI
     • Spam/Archive

3. Summarize

   • Each thread condensed into 2–3 clear bullet points so I instantly know what’s important.

4. Act Automatically

   • Draft personalized replies (not just templates) for urgent items.
   • Create calendar events or to-do tasks from actionable emails.
   • Auto-archive clutter.

5. Daily Briefing

   • Every morning, a Slack/Teams message with:

     • “Here’s what to answer first”
     • Upcoming deadlines
     • Any follow-ups due today

Why It Matters

• Saves hours every week — no more inbox triage marathons.
• Prevents costly misses — important messages always float to the top.
• Mental clarity restored — finally achieve inbox zero without lifting a finger.

Creative Edge

Unlike basic “filter and tag” tools, this agent understands context.

• Birthday wishes get priority over bulk newsletters.
• Job offers don’t get buried in “Promotions”.
• Replies feel natural, using the tone and details from the ongoing conversation.

[jokek40702 (vashu mish)]

very well explained! Your idea is bursting with creativity and originality — I can’t wait to see it in action.

[technical-explorer-8 (Ishika Gagneja)]

AI “Code Review Bot”

Task

Reviews pull requests, examines code quality, and proposes enhancements prior to merging.

How It Works

1. Listens to GitHub webhook events.

2. Retrieves modified files through GitHub API.

3. Performs static analysis through SonarQube API.

4. Leverages OpenAI API to create improvement proposals.

Real-Life Example

Dev commits code → AI reports unused imports, proposes renaming of variables for readability, and identifies missing null check.

Result

1. Improved code.

2. Less bugs in production.

3. Fewer senior dev review hours.

[altimetry-specialist (Kashish Gupta)]

AI “Coding Debugger Sidekick”

Task
Monitors code, detects errors, finds solutions, and suggests fixes.

How It Works

1. IDE plugin detects errors in code.

2. Sends error logs to AI via Postman API call.

3. Searches Stack Overflow API and GitHub Issues API.

4. AI merges relevant solutions and adapts them.

5. Suggests fix in IDE.

Real-Life Example
You hit a NullPointerException in Java. AI instantly finds 3 Stack Overflow answers, adapts one to your context, and inserts a fix suggestion in your editor.

Result

Debug time cut by 50%.

Learning as you fix.

Reduces project delays.

[yonandualem (Yonas Andualem)]

EventOps AI Agent, End to End Event Automation

If I could hand one task to an AI agent, it would be running all logistics for my student tech events from start to finish.

• You enter the event theme and date
• The agent writes the agenda and talk title
• The agent generates a banner using a design API
• The agent posts the banner and registration link in Slack
• The agent creates a calendar event for the team and speaker
• The agent sends email invites using a mail API
• The agent posts a reminder 24 hours before the event in Slack
• The agent sends a feedback survey link to attendees

APIs and tools used:
• Notion API to create the agenda and checklist
• Bannerbear API to generate the event banner
• Slack Incoming Webhooks to post announcements and reminders
• Google Calendar API to create and share the event
• Mailgun API to send email invites
• Typeform API to send the feedback survey

Postman Notebook details:
The Notebook contains both the explanation of the workflow and the complete runnable Postman Collection. Each step is a request that can be executed directly inside the Notebook. The requests are arranged in the same order as the agent’s process, with text sections in between to explain what each step does and the variables it uses. Environment variables store event details, API keys, and links. This design allows anyone to import the Notebook, set their variables, and run the entire event workflow end to end without switching between files.

Notebook link: EventOps AI Agent, End to End Event Automation

[anticomplan (Anirudh)]

---

The AI Life Admin Ninja – My Bureaucracy Assassin

Tired of wasting hours on bills, renewals, and endless forms? AI Life Admin Ninja detects tasks from my inbox, pulls required details from secure storage, and completes them via the right APIs — from passport renewals to bill payments — then notifies me when it’s done.

How it works:

1. Detect tasks via email or scanned documents (Gmail API + OCR).

2. Classify & plan actions using AI/NLP.

3. Execute through relevant APIs (Gov, bank, utilities).

4. Notify me via SMS or chat.

Why it’s a game-changer:

• Runs completely in the background.

• Works across multiple services without manual input.

• Turns hours of admin work into a single automated workflow.

With Postman orchestrating the workflow, my life admin now runs quietly in the background while I get on with living.

[vidhyalakshmin658 (vidhya N)]

AI “Legacy Code Whisperer”

The Problem

Every software team has that one mystery system:

• Built years ago by devs who’ve since moved on.

• APIs are undocumented, outdated, or both.

• Any change feels risky because no one knows what might break.

• Onboarding new engineers takes weeks just to understand it.

The result? Slower delivery, unexpected outages, and countless wasted engineering hours.

---

The Solution
The Legacy Code Whisperer AI agent:

• Discovers all reachable endpoints — even undocumented ones — via API crawling + traffic log analysis.

• Documents each in plain English with sample requests, responses, and usage notes.

• Scores fragility based on error rates, last update time, and dependency depth.

• Monitors continuously for schema or performance changes, delivering weekly summaries.

• Generates onboarding packs so new developers can start contributing in days, not weeks.

---

What We Achieve

• Faster onboarding → Days instead of weeks.

• Reduced risk → Heatmaps highlight fragile areas before changes.

• Better documentation → No more hunting through stale Confluence pages.

• Proactive alerts → Catch issues before they hit production.

• Time savings → Seniors focus on building, not endless hand-holding.

Notebook link: Legacy Code Whisperer

[navigation-observe27 (Dhanvantari)]

The Ultimate Coffee-First Morning Prep Agent

Because my brain doesn’t clock in until the caffeine does.

---

The Struggle

I’m not a morning person.

The first 30 minutes of my day? Mostly me staring at my phone like it’s speaking ancient Greek, while my coffee machine works in slow motion.

So why not outsource my entire morning brain to an AI agent that kicks in before my caffeine does?

How My AI Morning Butler Works

Step 1 – Wake-up Trigger At my alarm time, the AI wakes up before I do. It’s an overachiever.

Step 2 – Coffee Command Calls my smart coffee machine API to immediately start brewing espresso. Because priorities.

Step 3 – Overnight News & Weather Digest Pulls fresh headlines from NewsAPI
Fetches the local forecast from OpenWeather API
Result: A 30-second, no-fluff morning bulletin.

Step 4 – Calendar & Commute Check Reads my Google Calendar for today’s events.
Checks Google Maps API for traffic so I get the real leave-by time. No optimistic lies.

Step 5 – Priority Email Scan Uses Gmail API to read unread messages.
Flags only urgent and actionable ones.
Bonus: Anything from my boss gets a red alert.

Step 6 – Fun Boost Pulls a random joke or fact from a trivia API.
Because mornings need smiles, not stress.

Step 7 – The Delivery Packages all this into a neat, bullet-point Morning Report.
Sends it to me via Telegram Bot API or Slack API.
Perfectly timed to arrive with my first sip of coffee.

APIs and Tools My AI Agent Uses

• Smart Coffee Machine API (Smarter Coffee / IoT control)

• NewsAPI

• OpenWeather API

• Google Calendar API

• Google Maps API

• Gmail API

• OpenAI API (for summaries & tone)

• Telegram Bot API / Slack API

Why This is My Dream Agent

It turns my mornings from “ugh” to “ahhh”.

I don’t touch my phone.
I don’t think.
I just sip coffee while my AI has already:

Made my espresso
Briefed me on the day
Saved me from reading 20 irrelevant emails
Made me smile before 8 AM.

---

[flight-astronaut-586 (Shivam Tyagi)]

Shivam Tyagi – CSE BTech Student | Creator of 8tx AI

Hi, I’m Shivam Tyagi, a Computer Science & Engineering BTech student with a passion for AI, cybersecurity, and human-computer interaction.

I’ve always been fascinated by the idea of creating a truly personal AI — not just a chatbot, but something that could think, act, and even look like me.

That’s how I came up with 8tx — my dream AI agent and digital twin.

The Vision:

8tx is a complete virtual version of me — it talks like me, thinks like me, reacts like me, and even has a virtual body that mirrors my expressions and movements in real time.

Core Capabilities:

Human-Like Interaction – Engages in natural conversations, detects emotions, and adapts tone & behaviour like a real human.

Full Virtual Assistant – Opens apps, controls keyboard & mouse, manages schedules, sends messages, and automates daily tasks.

Creative Intelligence – Uses computer vision for virtual painting, gesture control, and interactive 3D environments.

Virtual Body Clone – A digital avatar for meetings, streaming, or remote presence.

System Control & Automation – From launching programs to executing multi-step workflows instantly.

Cybersecurity Edge:

Cyber Deception Hunter – Detects advanced fake encryption messages designed to distract defenders, instantly identifying real threats.

Self-Learning Security – Continuously trains on both ethical & unethical hacking techniques for maximum defense.

Threat Prioritization – Focuses on genuine dangers while ignoring distractions.

→ Why it’s unique:

8tx isn’t just a chatbot or assistant.

It’s my digital twin — a productivity partner, creative companion, virtual presence, and cy

ber bodyguard all in one AI agent.

→Important thing must be noticed this name of my virtual ai is not permanent and it can be changed it is a temporary name which I used.

[rishabh1902 (Rishabh Kr. Pandey)]

Relentless Follow-Up Bot — Because Important Messages Deserve to Be Answered

As someone who’s watched important client emails and team messages vanish into the abyss of “I’ll reply later,” I know the silent chaos that follows — missed opportunities, awkward reminders, and projects stalling for no real reason.

That’s why my dream AI agent would be a Relentless Follow-Up Bot — a tireless digital assistant that scans my workspaces, finds conversations that slipped through the cracks, prioritizes them, and gently (or not-so-gently) reminds me until they’re done.

How it works:

• Smart Scan: Connects to Gmail, Slack, or project tools via APIs to detect unanswered messages, open threads, or pending approvals.

• Context Awareness: Reads the last exchange to understand urgency and importance — no spammy reminders for trivial chats.

• Daily Priority Digest: Sends a short, actionable list like:

  “3 follow-ups today: Client proposal sent 4 days ago, budget approval pending since Monday, design feedback due in 2 days.”

• Auto-Response Drafts: Suggests polite, relevant follow-up messages you can send instantly.

• Persistence Mode: Keeps tracking until the conversation is closed or acknowledged.

In short, my AI Relentless Follow-Up Bot would give every busy professional what they secretly need: the ability to never let a conversation die, without the mental load of remembering every loose end.

Postman Notebook: I created a simple Notebook to show my AI agent in action.
Relentless Follow-Up Bot Notebook

[spacecraft-geologi13 (Arya mishra)]

The AI Archivist: How AI agent Could Save Us from Digital Chaos

Introduction

Our digital lives are messy. From thousands of photos on our phones to scattered documents across drives, cloud storage, and email attachments, finding the right file at the right time often feels like hunting for a sock in a dryer that eats socks. While modern AI dazzles with chatbots and image generators, conventional AI—rule-based systems, pattern recognition, and structured data processing—still shines when it comes to one thing: order.

So, if I could hand one task over to an AI agent, it would be personal digital archiving—a tireless, meticulous system that organizes my entire digital footprint with clockwork precision.

---

The Task: AI-Powered Digital Archivist

Instead of relying on me to name files properly, create folders, or remember where I saved things, my AI Archivist would:

1. Ingest all my digital assets from multiple sources (Google Drive, Dropbox, local hard drive, emails, phone gallery).

2. Classify files using conventional AI methods—decision trees, keyword extraction, and metadata parsing.

3. Apply consistent naming conventions so every file is searchable.

4. Tag content with structured categories (project name, date, document type).

5. Detect duplicates and old versions to reduce clutter.

6. Suggest archiving or deletion for obsolete files.

The twist? It wouldn’t need deep learning GPUs—it could run efficiently on a laptop or small cloud instance using conventional AI algorithms combined with APIs for file handling and search indexing.

Why this?

Deep learning gets the headlines, but for structured tasks like this, rules + statistical models + metadata analysis are:

• Faster to develop (no massive dataset required)

• Easier to interpret (clear decision rules)

• More efficient (can run locally without huge compute)

By combining metadata parsing with a set of deterministic rules—such as naming formats or file type handling—we get predictable, repeatable results.

---

Example AI Archivist Flow (Postman Notebook Ready)

Step 1: Connect to cloud storage API (e.g., Google Drive API) to list files.
Step 2: Fetch file metadata (title, type, created date, last modified).
Step 3: Apply classification rules:

if "invoice" in filename or OCR text → Tag as Finance
if filetype == "jpg" and contains faces → Tag as Personal Photo
if last_modified < 2018 → Tag as Archive Candidate

Step 4: Rename and organize using API calls (move to correct folder).
Step 5: Update an index for quick search.

---

Why I’d Love This

Instead of spending Sunday afternoons cleaning up my computer, I could focus on actually creating things, while my AI Archivist quietly keeps my digital universe tidy. It’s not flashy like generating a Picasso-style self-portrait—but it’s the kind of invisible magic that makes life immeasurably better.
Notebook: Postman

[track-baby-101 (Ogagaoghene Esavwede)]

The AI Agent That Knows Exactly Which Tests to Run (So I Don’t Have To)

If there’s one thing that always slows me down as a developer, it’s figuring out which tests to run after making a change

• Add a new feature? I have to think about unit tests, integration tests, maybe an end-to-end run.
• Change an API endpoint? Suddenly I’m worrying about smoke tests, load tests, and performance tests.
• Remove old code? Now I’m wondering which tests even matter anymore.
• And worst, for me, actually writing tests robust enough to cover all scenarios and edge cases.

And that’s before I actually write or run them.

So here’s my dream AI agent Task:

The Auto Test Strategist & Executioner

Step 1: It notices my code changes

The moment I implement a code or feature locally and also when I push code to GitHub, this agent looks at the diff. It understands what was added, changed, or removed, right down to the function level.

Step 2: It figures out the right tests to run

It knows that:

• A small function change might only need unit tests.
• A DB schema change probably needs integration and performance tests.
• A change to the checkout flow? Definitely end-to-end tests.
• New deployment scripts? Time for smoke tests.

Step 3: It writes missing tests for me, if any

If there’s a gap like a function with no coverage it drafts a test case, in my preferred framework, so I can review and approve.

Step 4: It runs them locally in development and in the cloud when pushed

The tests are run locally before leaving my machine, and also in the cloud as part of the cicd pipeline and sends me
results

Step 5: It reports like a pro

Instead of a boring “passed/failed” message, I get:

• Test coverage % for this commit.
• The slowest 10 tests.
• Any flaky tests detected.
• Performance trends over the last 10 runs.

Why I’d Love This Agent

It’s like having an expert QA lead on my team, 24/7, who already knows my codebase and testing strategy. I could focus on building features instead of babysitting tests and my confidence in every deploy would skyrocket, as well as my peace of mind.

I think it’s the kind of thing every developer wishes they had and my AI agent would finally make it real.

[avionics-cosmologis2 (Chinakshu Mangal)]

Project: The Context Keeper

What It Is: An AI agent that acts as your perfect digital memory, remembering every detail from your meetings, calls, and emails so you don’t have to.

The Problem It Solves:

• Forgetting key decisions and action items from past conversations.

• Wasting the first 10 minutes of meetings just catching up.

• Missing opportunities to build stronger relationships by forgetting personal details.

• Feeling overwhelmed by information overload.

How It Works:

1. Listens & Reads: Securely ingests information from your calendar, calls, and messages.

2. Understands & Summarizes: Uses AI to identify key points, tasks, and personal notes.

3. Reminds You: Sends you a simple, powerful briefing before your next meeting.

Example Pre-Meeting Briefing:

Upcoming Meeting: Priya (3:00 PM)

• Last Topic: Discussed new website design; Priya wanted a different logo.

• Your Task: Send 3 new logo options. (Done)

• Priya’s Task: Send text for the “About Us” page.

• Personal Note: Ask how her son is doing with his guitar lessons.

The Bottom Line: The Context Keeper’s goal is to make you seem more prepared, attentive, and organized with zero extra effort. It handles the remembering so you can focus on the conversation.

---

Postman Notebook: See It in Action

To see a simplified technical sketch of how this AI agent would use APIs to gather, process, and present information, check out the Postman Notebook below.

Context Keeper AI Agent - Notebook

[shaiquaperween1428 (Shaiqua Perween)]

“My Dream AI Agent: The ‘Life Admin Ninja’”

If I could hand one task over to an AI agent, it would be managing all the small, boring, repetitive life chores that eat up my mental energy — but in a proactive way.

Here’s what my AI Life Admin Ninja would do:

• Email & Message Filter → Automatically read, summarize, and prioritize my inbox and chats, so I only see the important ones (and get the 2-line summary instead of 200-line threads).

• Bill & Deadline Guardian → Track my bills, subscriptions, and project deadlines, pay/submit them automatically, and warn me before something expires or is due.

• Smart Calendar Curator → Not just add events, but schedule my day with balanced work, breaks, and even “fun time” based on my mood, deadlines, and energy levels.

• Life Memory Vault → Remember where I saved that one file from 6 months ago, or that API endpoint I once tested — and pull it instantly when I ask.

• Decision Simplifier → When I’m stuck choosing between 10 similar products, jobs, or courses, it compares all the data, reviews, and my preferences to recommend the best fit.

The twist?
It would work silently in the background like a personal assistant who knows me better than I know myself, freeing me to focus on creating, learning, and living instead of drowning in “digital paperwork.”

[gbadebo-bello]

That’s a wrap! This week’s challenge is officially closed. Thanks to everyone who jumped in.

We’re reviewing submissions now.

While you wait, drop a on your favorite responses to help us choose a winner.

Winner will be announced on Friday.