Documents

Selectable regions drawn over a page image from normalized coordinates, for showing an agent exactly where an answer came from.

Notes attached to regions of a page. Drag to add one, select to read it, and the coordinates stay relative to the page rather than the screen.

Mark regions to black out before a document leaves the building, with a reveal that says plainly it is only a preview.

A rail of page thumbnails for moving through a long document, with arrow-key navigation and a clear active page.

An expandable tree of folders and files with full keyboard navigation and correct tree semantics.

Mark where one scanned batch becomes several documents. Splits are toggled between pages and the segments update as you go.

Build the shape you want extracted from a document. Fields carry a name, type, description, and requirement, and object and array fields nest.

Sign with a pointer on a canvas, or type a name instead. Returns a PNG data URL or the typed text.

A real table for delimited data, with sortable columns, a sticky header, and a row cap so a large file cannot lock the page.

A collapsible tree for a JSON payload, navigable from the keyboard, where every row can hand you its path.

Renders a Word document as elements built through an allowlist, so a file you did not write cannot bring its own scripts or links.

Page-by-page PDF rendering on a canvas, with paging and zoom, over any loader you give it.

Extracted document regions rendered as markdown, each one selectable so it can be tied back to where it came from.