About AccuraCite

AccuraCite started with a bibliography I didn't write myself. When you draft your own citations, you know where each one came from. But I was co-authoring a paper where part of it was AI-assisted, and the citations looked fine on the surface until I actually checked them. Some misquoted the source, and a few didn't exist at all. That's the real risk: you can vouch for what you verify yourself, but not for what someone else hands you. And a fabricated citation isn't just an embarrassing mistake to catch before submitting: if a reader goes looking for that source, it isn't there, and the knowledge they were trying to find never existed in the first place. Checking a full reference list by hand is slow, so I built a tool that checks every citation against real databases (OpenAlex, Crossref, Semantic Scholar, PubMed, DBLP, arXiv) and tells you when something's wrong before you hit submit.

The source-finding tool is for a related reason. A paper's claims need real sources behind them, not just something in the right subject area. I've read plenty of papers with reference lists that were clearly padded to boost a citation count or satisfy a reviewer. The citation was there, but it didn't actually back the claim. AccuraCite's source-finding is built to make finding a citation that genuinely supports your claim just as fast as reaching for one that merely sounds related.

I have a PhD in computer science and worked as a software engineer before that. AccuraCite is a solo project. I built it for my own work first, and kept building it after other people started asking for it.

Get in touch via the Feedback page with questions, bug reports, or ideas.