RECLaiM

Inspiration

As individuals with ADHD, we’ve often struggled to stay focused while studying online. Existing tools like website blockers felt overly restrictive, often blocking entire sites we needed access to, leading to frustration and frequent disabling of the tools. This inspired us to create a smarter solution: RECLaiM, an adaptive Chrome extension that uses AI to dynamically evaluate website relevance based on the user's focus, providing a tailored approach to staying on track.

What it does

RECLaiM is a Chrome extension designed to enhance focus during study sessions. Users begin by specifying their study topic and method, and RECLaiM leverages large language models to analyze the content of visited websites. If a site’s content doesn't align with the user’s focus, it is dynamically blocked, presenting a prompt with the reason for the block. Users can justify access to the site if necessary, creating a flexible, intelligent study companion that encourages productivity while allowing room for legitimate diversions.

How we built it

We developed the frontend using React and TypeScript for a clean and responsive user interface. The backend, built with Flask, handles processing user inputs and analyzing website content using a large language model. Communication between the browser, service workers, content scripts, and backend was key to integrating all components. This architecture ensures seamless interaction and decision-making during study sessions. We developed a Google Cloud console infrastructure with Terraform for our backend and data storage.

Challenges we ran into

Building a Chrome extension for the first time presented unique challenges. We struggled with setting up efficient communication between service workers, content scripts, and the backend. Additionally, fine-tuning the LLM to accurately assess website relevance for diverse topics required thoughtful testing and adjustments. Assembling and debugging these interconnected systems was a steep learning curve but a rewarding experience.

Accomplishments that we're proud of

We are proud of creating a fully functional, end-to-end project within the hackathon timeframe. For two team members, this was their first software project outside of course work. The other two team members are more experienced with hackathons, but it was their first time building a Chrome extension. Successfully integrating AI into a practical tool and mastering new skills along the way was a significant achievement.

What we learned

This project taught us how to interface between multiple components of a Chrome extension, including frontend, backend, service workers, and content scripts. We also gained valuable experience working with large language models for unique use cases, balancing flexibility and strictness in content analysis to meet user needs. Collaborative problem-solving under tight deadlines was another critical takeaway.

What's next for RECLaiM

We plan to scale RECLaiM by refining its AI algorithms for greater accuracy and adding features like session analytics, personalized productivity tips, and advanced focus tracking. We also aim to publish the extension on the Chrome Web Store very soon, making it accessible to a broader audience and helping more users reclaim their focus with smarter, adaptive tools.