Connect a GitHub repo and get its estimated professional rebuild value in minutes. See what you’ve actually built, what makes it valuable, and what it would take to recreate it today.
Most of us have a folder of side projects we've poured years of evenings into, with no real idea whether we built something substantial or a pile of nothing.
So to help, I poured some more evenings into this.
Connect GitHub (readonly), pick a repo, and a couple of minutes later you get a number: what it would cost to commission a production-ready equivalent today. Free.
Most code tools count lines and call it insight. This works out what your software actually does, the capabilities, the integrations, the genuinely hard bits, and prices those. 20,000 lines of duplicated CRUD earns you nothing. 800 lines doing something properly difficult earns you a lot. It also assumes whoever rebuilds it has the same AI tools you do, which is the bit most estimates get wrong.
Honest bit: the first version valued itself at £99,600, which was nonsense. It was counting the same feature four times (frontend, API, data model, integration). That sent me down a proper rabbit hole and I rebuilt the methodology around what the product does instead of how it's layered.
£49 gets the full report: every capability with hours and cost, the actual files cited as evidence, an architecture grade, the technical debt it found, and what'd be hardest to reproduce.
Private repos work, read-only, source is never retained.
What I'd genuinely love: run something you know inside out and tell me if it's right or miles off. I'd rather hear "this is way too high" than believe my own tool.
Most of us have a folder of side projects we've poured years of evenings into, with no real idea whether we built something substantial or a pile of nothing.
So to help, I poured some more evenings into this.
Connect GitHub (readonly), pick a repo, and a couple of minutes later you get a number: what it would cost to commission a production-ready equivalent today. Free.
Most code tools count lines and call it insight. This works out what your software actually does, the capabilities, the integrations, the genuinely hard bits, and prices those. 20,000 lines of duplicated CRUD earns you nothing. 800 lines doing something properly difficult earns you a lot. It also assumes whoever rebuilds it has the same AI tools you do, which is the bit most estimates get wrong.
Honest bit: the first version valued itself at £99,600, which was nonsense. It was counting the same feature four times (frontend, API, data model, integration). That sent me down a proper rabbit hole and I rebuilt the methodology around what the product does instead of how it's layered.
£49 gets the full report: every capability with hours and cost, the actual files cited as evidence, an architecture grade, the technical debt it found, and what'd be hardest to reproduce.
Private repos work, read-only, source is never retained.
What I'd genuinely love: run something you know inside out and tell me if it's right or miles off. I'd rather hear "this is way too high" than believe my own tool.