One check per kind of output, all of them here. Run any of them and you get the specific problems, where each one is and the evidence for it, with the same verdict on every run. Writing starts on our own homepage copy, because we check our writing with the thing we sell.
Dead prose, hedging, jargon, self-management tics, flat cadence. Paste any AI text.
Upload a PDF. The extraction check detects the table structure and pulls the cells deterministically, the same numbers every run. Byte-identical across 90 runs, 2,332 cells audited with 0 errors. (For the full visual overlay on the source page, the extraction check also runs standalone at documents.doloop.io.)
Paste a link. The website check crawls the site and runs all four faces in one pass: Structure (one header, one menu, no orphans), Voice (AI-writing tells), Words (a caged reader on the headlines and copy), and Design (the CSS audit plus a vision pass on the layout). Same site in, same verdict out.
Paste code. The check flags security gaps (a network length reaching a copy with no bounds-check, CWE-125), smells (long parameter lists, swallowed errors), hardcoded secrets, and documentation drift (a docstring naming a parameter the function does not have). Same code in, the same verdict out, no model.
The richer demos run on their own pages. Each one is live, deterministic, and free to try.
DOMAINS reads any draft across seven dimensions of voice and shows the one change toward the voice you want. Same read every run.
Pull a table out of a real PDF and click any value to see the source page render with that word boxed. Located, not generated.
Drop in a public GitHub repo and get its map in a minute: the load-bearing files, the backbone, where a change ripples. No install.
A 60-second self-test for whether you can trust what your model just extracted. Four questions that make it tell on itself.
Paste your URL and watch a bot get built on your own pages. It answers from your content or says it does not know. No model on the answer path.
Paste your sitemap and doloop reads every page live: the headers that drifted, the pages with no link in, the voice that turned inward. We ran it on this site first.
Routes you to a real doloop page, asks when your question is ambiguous, or tells you when there is no answer. No model runs on the answer path, so it cannot invent one.