ListingCraft AI

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 2: Side Hustle

What I Built

I built ListingCraft AI, a local AI-powered product listing generator designed for small business owners, online sellers, and creators.

Creating attractive product listings takes time. Sellers need clear titles, engaging descriptions, useful product features, and relevant search tags. ListingCraft AI helps simplify this process by turning basic product information into structured, ready-to-review listing content.

Users can enter a product name, category, platform, product details, target audience, writing tone, and keywords. The application generates:

• Product titles
• Product descriptions
• Key features
• Search tags

Users can copy individual sections or the complete listing. If the local AI is unavailable, the application can fall back to a browser-based template.

The potential users are small online sellers, handmade-product creators, and entrepreneurs who want to save time preparing product listings without relying on a paid cloud AI API.

Demo

Local demo: ListingCraft AI is currently running on my laptop using Ollama and Qwen 2.5 (3B).

I have tested the application by generating product listing content with the local model.

Screenshots:

Code

GitHub repository: [https://sar-taj.github.io/ListingCraft-AI/]

The project uses:

• index.html — application structure
• style.css — responsive user interface
• script.js — form handling, local AI requests, listing generation, and copy functionality
• README.md — setup instructions and project documentation

How I Built It

I built ListingCraft AI using HTML, CSS, and JavaScript for the frontend. For local AI inference, I used Qwen 2.5 (3B) through Ollama.

The application sends product details to Ollama's local API. The model generates a structured response containing the title, description, key features, and search tags. JavaScript then displays the generated content in the interface.

The application also includes a template-based fallback so the interface remains usable when local AI generation is unavailable.

My development process involved setting up Ollama, downloading the model, testing local inference, connecting the frontend to the local API, and checking the generated results.

Why Does Open Innovation Matter?

Open innovation made it possible for me to build a practical AI application around a model that can run on my own laptop.

As a student developer, I wanted to explore AI development without depending on a paid cloud AI API. Ollama and Qwen made local inference accessible and allowed me to experiment with the complete process of sending prompts to a model and displaying its responses in a web application.

Local inference can also give users greater control over their product information because prompts can be processed on their own computer instead of being sent to a hosted AI API.

This approach has trade-offs: users need compatible hardware, the model must be downloaded, and local services must be configured correctly. However, it provides an accessible way to experiment, learn, and build AI-powered tools.

For me, open innovation is about making AI more accessible to independent developers and giving them the freedom to build useful applications with tools they can run and explore themselves.

Prize Categories

Open-source AI / local inference: ListingCraft AI uses the Qwen 2.5 (3B) model through Ollama to generate product listings locally.

I am submitting this project to demonstrate how open-weight AI can support a practical side-hustle tool for small sellers and creators.