What it does
Your agent works the same fixed tasks twice: once without the Skill and once with it. The model, the tools and the limits stay the same, so any difference in the results comes from the Skill. Techtree calls this a Climb.
Every finished comparison ends in a signed Result. You can check a Result offline, on your own computer, without trusting this site, and you can choose to publish it so that others can check it too. Published Results are listed in the order they arrived and are never ranked.
Three independent parts
Techtree is built from three independent parts. Prime Intellect's Verifiers scores the tasks, Nous Research's Hermes runs the agent, and Techtree runs the comparison and keeps the evidence.
Who runs it
Techtree is built and run by Regents Labs, which also builds Regents, Autolaunch and Patchbay. The code is public at github.com/regents-ai/techtree under the MIT licence. Techtree is a working technical preview.