opd, the Open Page Document
In build
An open file format meant to replace PDF in the common case, which is showing a fixed, formatted document without the weight of the PDF toolchain. One .opd file is one zip: a clean, reflowable source in a strict subset of HTML and CSS with no scripting and nothing loaded from outside, the fonts, and optionally a frozen copy of the exact pages, one SVG per page, bound to the source by hash. Around it sit a viewer, an editor and a PDF importer that are each a single HTML file and work offline from any folder. The repository went from empty to all of that between October 16 and October 19, 2026.
- No modes: every valid file opens editable under one standard toolbar, a form opens ready to fill inside the editor, and the editor can only produce valid files because its schema is generated from the format's profile
- Create, send, fill, return: filled answers live apart from the author's bytes, a form fingerprint shows a copy came back unaltered, Ed25519 signatures are worded to claim exactly what they prove, and a Collect view exports returned copies as CSV on the device
- PDF import scored against a 35-document ground truth rather than eyeballed: reading order 1.00, exact text 0.988, nine tables in ten, and an optional AI pass reported honestly as no measured gain
- Built by a Board of Claude Code subagents in which only the Engineer writes code, through about 55 pull requests, with more than 1,000 unit tests, browser tests in Chromium, Firefox and WebKit, and accessibility checks held at zero findings


