Mochi

• 

  Choose how you read: simplification level, reading mode, and dyslexia-friendly display options — all in one place.

• 

  Fine-tune your reading experience with spacing controls and simplify any page in a single click.

Mochi-Web — About the Project Inspiration

The web is loud.

Big paragraphs, complex language, and zero consideration for how different people process information. For users with dyslexia, attention difficulties, or cognitive overload, reading the web can be unnecessarily exhausting.

Most accessibility tools either oversimplify content, remove important context, or are hidden deep inside browser settings that users rarely touch.

We wanted something simple:

Click once, and the web becomes readable.

That idea became Mochi-Web — a tool designed to soften the web and reduce reading friction.

What it does

Mochi-Web is a Chrome extension that uses AI to simplify web content while preserving its original meaning.

It allows users to:

Choose a simplification level (Low, Mid, High)

Select a reading mode (General, Focus, Processing)

Customize their reading experience with:

Open Dyslexic font support

Adjustable line height, letter spacing, and word spacing

Multiple themes including dark, sepia, cream, and high-contrast modes.

How we built it

Mochi-Web is built as a Manifest V3 Chrome extension with a modular architecture:

Popup UI for user controls and preferences

Content scripts for extracting and replacing page content

A service worker to handle asynchronous API calls

Gemini API for AI-powered text simplification

Workflow:

User clicks “Simplify Page”

Page text is extracted by the content script

Text is sent to the service worker

The service worker calls the Gemini API

Simplified content is returned and rendered on the page

User preferences are stored locally using Chrome Storage.

Challenges we ran into

Preserving meaning while simplifying text without making it inaccurate or vague

Handling Chrome Manifest V3 service worker lifecycle and message timing

Designing reading modes that work for different cognitive needs

Managing API key security for a browser-based AI tool

Accomplishments that we're proud of

Three distinct simplification levels that feel meaningfully different

Reading modes designed for focus and processing needs

Real-time customization of fonts, spacing, and themes

One-click restoration of original page content

A project focused on real accessibility problems, not just features

What we learned

Accessibility requires intentional design, not afterthoughts

Small UI and typography changes can significantly reduce cognitive strain

AI is most effective when it adapts to user needs

Chrome extensions require careful handling of asynchronous behavior

What's next for Mochi-Web

Planned improvements include:

Secure API key management through a backend or user-provided keys

Selection-based text simplification instead of full-page replacement

Publishing to the Chrome Web Store

Multi-language support

Performance and offline-friendly enhancements

Long-term, Mochi-Web aims to help create a web that adapts to the reader instead of forcing the reader to adapt to the web

Built With

• css3
• html5
• javascript