Scan a board off your screen
Drag a selection around the 64 squares. That single crop is the only thing sent for recognition — manual moves are never streamed anywhere.
47 Apps · the software side of Forty Seven Lab
47 Apps is where ideas inside Forty Seven Lab become things you can actually run. Tools built around problems we keep running into, experiments that occasionally turn into products, and the odd piece of software that simply should have existed already.
4 projects in development right now
Featured application
A Windows chess companion that reads a position straight off your screen, rebuilds it exactly, and hands you a real board — to play on, or to analyse with a local engine.



Every image here is a real capture of Chess Vision 3.5, not a mock-up.
Drag a selection around the 64 squares. That single crop is the only thing sent for recognition — manual moves are never streamed anywhere.
Recognition returns all 64 squares as structured data, and the result is validated before it is accepted. Missing or impossible squares are rejected rather than guessed at.
Analysis for unrestricted work on a position, local two-player, and play against ChessAI at an approximate 600–2800 target strength.
Local Stockfish supplies a best-move arrow, move notation and a live evaluation bar. Suggestions are a preference, not a permanent overlay.
1, 5, 10 or 20 minutes per side with timeout adjudication, player names, and captured-piece displays that follow the board's orientation.
Left and right arrows step through the recorded history — turn, castling rights and en-passant state included, not just a picture of a board.
Chess Vision has its own showcase — the scan flow, every game mode, the analysis tools and what is under the hood, all with real screenshots.
Open the Chess Vision showcaseMore in the lab
These are the slots behind it — work in progress, in the order it is actually happening. None of them have names yet, because naming something before it exists is how you end up shipping a logo instead of a product.
Past the sketch stage and being built properly, with a release date we are not going to promise yet.
Working enough to be useful to us. Whether it becomes a product depends on whether it survives contact with real use.
An idea being poked at in the lab. It might turn into something. It might quietly disappear. Both are fine.
Finished, packaged and waiting for the right moment to be made public.
We would rather show you one finished application and four honest placeholders than a wall of logos for things that do not exist. When something in here is ready, it moves up and gets a real page.
About 47 Apps
47 Apps is the software division of Forty Seven Lab. One application is public, more are in the workshop, and the standard for both is the same: it has to be worth the disk space.
Some of what we build starts as a client problem, some as an internal tool, and some simply as a disagreement with how something already works. What they have in common is that we wanted them to exist and could not find them elsewhere.
That means the portfolio will always look a little uneven — a chess companion next to an accounting tool next to something that only makes sense once you have seen it. We are comfortable with that. A software studio that only builds things which fit one category is a marketing department with a build step.
Currently4 projects in the workshop, one public release, and an open door for ideas that belong to somebody else.
A tool you reach for twice a week is worth more than a demo that photographs well. Every app here starts from something that annoyed us enough to fix.
Desktop where desktop is right, local where privacy matters, and no dependency we cannot explain. If we cannot maintain it, we do not ship it.
Chess Vision is released under the GPL with its full source public. Reading the code is allowed to be part of using the software.
Build an idea
Almost everything on this page started as somebody saying “why isn’t there a tool that just does this?” If you have an idea sitting in your head — an app, an internal tool, a process that should run itself — bring it to the lab. We will tell you honestly whether it is worth building, and whether we are the right people to build it.
No pitch decks, no discovery-call funnel. A conversation first.