DEV Community: Shashank Dixit

GHOSTWRITE AI: The Haunted Storyteller — Where Artificial Intelligence Meets the Supernatural.

Shashank Dixit — Fri, 30 Oct 2026 04:52:13 +0000

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 4: Ghost in the Machine

What I Built

I built GHOSTWRITE AI — The Haunted Storyteller, a horror storytelling web application inspired by the theme "Ghost in the Machine."

The project explores the mysterious side of artificial intelligence through a spooky interface where users can create horror stories, interact with a fictional haunted terminal, and save their favorite stories in a Memory Garden.

My goal was to combine creative storytelling, web development, and AI into an interactive experience that makes technology feel a little supernatural.

Code

GitHub Repository: https://github.com/shshnkdixit/GHOSTWRITE-AI

The project is built with HTML, CSS, and JavaScript.

Demo

Live Website: https://shshnkdixit.github.io/GHOSTWRITE-AI/

How I Built It

I developed GHOSTWRITE AI using HTML for the structure, CSS for the dark and atmospheric design, and JavaScript for the interactive features.

I used ChatGPT as a development assistant to help write code, improve the interface, and organize the project.

The project is designed to support local AI integration using Ollama and an open-weight model such as Llama 3.2. This approach can enable story generation without relying entirely on a closed AI API.

The frontend is hosted on GitHub Pages.

Why Does Open Innovation Matter?

Open innovation gives developers the freedom to experiment, learn, customize tools, and build applications without depending entirely on proprietary platforms.

For this project, open-weight AI models offer an opportunity to explore local inference, experiment with creative applications, and maintain greater control over how AI-generated content is produced.

As an aspiring AI engineer, this project helped me explore how AI can be combined with web technologies to create something interactive and creative.

My Agent Session

This section is optional. I have not included an agent session because I do not have a DevRelay session link to share.

The Spirit of It

What if a machine could remember a story long after its creator had forgotten it?

GHOSTWRITE AI explores that idea through horror storytelling, mysterious terminal messages, and a Memory Garden where stories can be preserved.

The project brings the theme "Ghost in the Machine" to life by blending the eerie atmosphere of supernatural fiction with the possibilities of artificial intelligence.

Prize Categories

My project focuses on open-source AI and creative applications built around the "Ghost in the Machine" theme.

SlothFlow — Your Lazy-Day AI Assistant

Shashank Dixit — Wed, 21 Oct 2026 07:38:09 +0000

This is a submission for the Hacktoberfest Open-Source AI Challenge Week 3: Slow and Steady

What I Built

I built SlothFlow — Do Less, Live More, a privacy-focused AI productivity assistant designed to make everyday work simpler and less overwhelming.

The idea behind SlothFlow is straightforward: productivity should not mean constantly feeling busy. Sometimes, the hardest part of getting things done is figuring out where to start.

SlothFlow helps users turn unstructured thoughts and to-do lists into manageable action plans, summarize notes, and prioritize tasks. It also includes a task management dashboard where users can organize their work and track their progress.

The project includes three main AI tools:

Task Planning: Break overwhelming tasks into smaller, actionable steps.
Smart Summarization: Turn lengthy notes into concise, useful summaries.
Task Prioritization: Identify important tasks and decide what deserves attention first.

SlothFlow also supports task tracking, browser-based persistence, and a demo mode so users can explore its features without setting up an AI model.

My goal was to build something that reduces the mental effort of organizing work and helps people focus on one meaningful step at a time.

Demo

Try SlothFlow here:

Live Demo — SlothFlow

The application features a calming, nature-inspired interface designed to make productivity feel less stressful.

Code

Explore the complete source code on GitHub:

GitHub Repository — SlothFlow

The project is built with HTML, CSS, and JavaScript, with optional local AI integration through Ollama.

How I Built It

I built SlothFlow using HTML, CSS, and vanilla JavaScript to keep the application lightweight, accessible, and easy to run.

The project combines a responsive productivity dashboard, browser-based task storage, interactive task management, and AI-powered productivity workflows.

For AI integration, SlothFlow supports Ollama, which allows users to run open-weight language models locally instead of relying exclusively on a hosted AI API.

For example, users can download a model such as Llama 3.2 through Ollama and configure SlothFlow to connect to their local model endpoint. The application can then send prompts to the locally running model for task planning, summarization, and prioritization.

The project also includes a demo mode, allowing users to explore the interface without installing a model.

My approach was to keep the core application simple while making local AI an optional capability. This makes SlothFlow easier to experiment with and gives users more control over how they use AI.

Why Does Open Innovation Matter?

I believe AI should be accessible, adaptable, and respectful of user privacy.

Many AI productivity applications depend on centralized services, which can require users to send personal notes, plans, and tasks to external servers. That is not always ideal, especially when the information is private.

Open-weight models and local inference provide another option.

By supporting Ollama, SlothFlow gives users the opportunity to run AI on their own machines, choose compatible models, and experiment without requiring a hosted AI API key for the local workflow.

Open innovation also makes the project easier to learn from and extend. Developers can explore the source code, adapt the interface, experiment with different models, and contribute new features.

For me, the most important benefit is user control. AI should help people get things done without forcing them to give up control over their data or depend entirely on a single provider.

SlothFlow is my small step toward making useful AI tools more approachable, private, and accessible.

My Agent Session

I built SlothFlow with a focus on keeping the application lightweight and understandable while integrating optional local AI capabilities.

I have not included a DevRelay agent session link in this submission. If I publish a session later, I can add it here.

The Lazy Payoff

SlothFlow is designed to reduce the time and mental effort spent deciding what to do, organizing notes, and breaking large tasks into smaller steps.

Instead of staring at a long to-do list and wondering where to begin, users can use the planning and prioritization tools to create a clearer starting point.

The goal is not to automate every part of productivity. It is to remove some of the friction that makes getting started difficult.

I am continuing to improve the project, with the aim of making everyday planning feel simpler and more manageable.

Prize Categories

My primary focus is the Slow and Steady open-source AI challenge: building a useful application around open-weight AI and local inference.

I have not selected any additional partner prize categories. I will update this section if I enter a specific eligible category.

Project: SlothFlow

Live Demo: https://shshnkdixit.github.io/SlothFlow/#ai-tools

Source Code: https://github.com/shshnkdixit/SlothFlow

LocalBiz AI: An Open-Source AI Side Hustle Assistant for Small Businesses | Hacktoberfest 2026

Shashank Dixit — Wed, 14 Oct 2026 09:29:10 +0000

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

What I Built

I built LocalBiz AI — Your AI Side Hustle Assistant, an open-source web application that helps small business owners, freelancers, and aspiring entrepreneurs create business content more easily.

Creating professional marketing content can be time-consuming, especially for people who are running a small business with limited resources. LocalBiz AI helps users generate product descriptions, social media captions, business growth ideas, and pricing suggestions from a simple dashboard.

Users can customize their content according to their business, target audience, and preferred writing tone. The application also supports saving previous creations and copying generated content.

The project is designed for small business owners, freelancers, students, and people starting a side hustle. Potential customers could pay for additional features such as advanced content generation, multilingual support, downloadable reports, and business-specific AI assistance.

Demo

Live Demo: Try LocalBiz AI

GitHub Repository: Explore the source code

Code

The project is open source and built with HTML, CSS, and JavaScript, with optional local AI integration through Ollama.

Repository: https://github.com/shshnkdixit/LocalBiz-AI

The code is organized into separate files for the user interface, styling, and application logic, making it easier for contributors to understand and improve.

How I Built It

live Url = https://shshnkdixit.github.io/LocalBiz-AI/

I built LocalBiz AI using:

HTML5 for the application structure.
CSS3 for the responsive interface and dashboard design.
JavaScript for form handling, content generation, copy functionality, and saving previous creations.
LocalStorage for keeping saved content in the browser.
Ollama for optional local inference using open-weight language models.

The application includes a demo generation mode and an Ollama integration path. Demo mode helps users explore the interface, while the Ollama integration allows the application to send prompts to a locally running language model.

I focused on making the interface simple enough for non-technical users. Instead of requiring users to write complex prompts, the application guides them through entering their business details and selecting the type of content they need.

Why Does Open Innovation Matter?

Open innovation makes AI more accessible to developers, students, and small businesses that may not have the budget for expensive proprietary AI services.

For LocalBiz AI, open-weight models and local inference offer a way to experiment with AI without making a paid proprietary API a mandatory part of the application. Developers can inspect the code, customize prompts, experiment with different supported models, and contribute new features.

This approach also encourages learning and collaboration. Other developers can extend the project with additional languages, better content-generation workflows, improved accessibility, and integrations for real small-business needs.

My goal is to explore how open-source AI can help people turn their business ideas into practical tools.

My Agent Session

This project was developed through an iterative coding workflow involving AI-assisted development, implementation, and refinement.

I do not have a public DevRelay agent-session link to share yet.

Prize Categories

My project is submitted for the Hacktoberfest Open-Source AI Challenge — Week 2: Side Hustle.

Additional partner prize categories can be listed here if the project meets their specific eligibility requirements.

Built by a developer exploring open-source AI and practical tools for small businesses.

If LocalBiz AI sounds useful, check out the repository, try the live demo, and contribute ideas for making it better!