Visual Reading Layer

About

Visual Reading Layer turns dense nonfiction into a calmer, more visual reading experience — without sending your reading anywhere.

Press one button to extract the article from the page you are on, or use your selected text, and read it in a focused reader with:

- Three reading surfaces: Clean, Emoji Rich, and Ultra Visual, with concept highlights and emoji acting as visual anchors, never replacing the original text.
- Insight panels: concept chips, cause-effect maps, contrast cards, an attention heat map, and an action board, generated by an inspectable rule engine that runs entirely in your browser.
- A local library: import DRM-free EPUB and TXT books on-device, archive extracted web articles, search your shelf, track per-book progress, and export notes metadata.
- Real reading tools: reading-progress restore, per-paragraph bookmarks, highlights and private notes, chapter navigation, Focus Mode, typography and color-theme controls, and keyboard shortcuts.
- Optional LLM enhancement: bring your own provider, including a local server, Google Gemini, DeepSeek, Xiaomi MiMo, or Z.AI GLM, for richer annotations, grounded insights, and bilingual translation.

Local-first by design: no backend, no account, no analytics, no tracking. The rule-based reader works completely offline. Optional LLM requests are strictly opt-in: nothing is sent until you enable LLM mode and press an action button, and only the paragraphs covered by that action are sent directly from your browser to your selected provider.