Textpilot

Inspiration

The idea for the Auto-Writing Assistant came from the need for a fast, reliable writing aid that can turn simple keywords into fully-formed paragraphs. In today's fast-paced world, content creators, students, and professionals often struggle with generating well-written text on demand. Inspired by tools like GitHub Copilot, which offers code suggestions, we wanted to create something similar for writers—a Chrome extension that could help users craft detailed paragraphs from just a few keywords. The aim was to build an assistant that boosts productivity, sparks creativity, and helps users overcome writer’s block.

What it does

The Auto-Writing Assistant allows users to enter a few keywords or phrases in a Chrome extension, then generates a complete, coherent paragraph based on those inputs. Users can adjust the tone (e.g., formal, casual) and length, making it adaptable for different content needs like social media posts, emails, articles, and more. With additional features like tone selection and alternative paragraph generation, it empowers users to generate text that is both relevant and well-structured.

How we built it

The project was built using:

JavaScript and HTML for the extension's interface, providing a simple and user-friendly popup in Chrome.
API Integration with Gemini neno (or similar), which handles the text generation based on user-provided keywords.
Manifest V3 setup for the Chrome extension, defining permissions and setting up a background script.
Message Passing between popup and background scripts for efficient API requests. The process involved designing a clean UI, handling API calls securely, and implementing asynchronous communication between components to ensure a smooth user experience.

Challenges we ran into

One of the main challenges was optimizing the prompt sent to the AI model to ensure relevant responses. Generating text from minimal keywords can sometimes produce vague results, so we iteratively refined the prompt structure. Another challenge was managing the Chrome extension’s message passing to keep it responsive, especially when dealing with API response times and network latency. Additionally, balancing tone and content while maintaining a high-quality output required careful tuning of the AI parameters.

Accomplishments that we're proud of

We’re proud to have developed a tool that genuinely assists users in content creation and improves productivity. Being able to turn a few words into a full, coherent paragraph was a satisfying achievement, and the extension's versatility is something we’re proud of. We also managed to implement tone selection, which adds an extra layer of customization for users.

What we learned

Building this project taught us a lot about integrating AI models into web extensions, managing asynchronous processes, and crafting meaningful prompts for content generation. We also gained experience in optimizing the Chrome extension workflow, learning how to balance functionality with user experience.

What's next for Textpilot

Our future plans include adding support for multilingual text generation, refining tone and style customization, and introducing features like summarization and keyword extraction. We’d also like to incorporate sentiment analysis so users can generate text that aligns better with the intended emotional tone. Additionally, we’re considering integrating cloud storage so users can save and retrieve their generated content directly from the extension.