Open Model Lab is being prepared as a public-source learning project, not a clone-and-deploy kit.

The goal is to make the learning model easier to inspect, improve, and learn from while keeping official domains, deployment config, vendor accounts, and private learner data out of the public tree.

Share a contribution idea

Current posture

Public-good learning first; optional Supporter features only fund sustainability.

Core concept pages, simulations, tools, guided paths, challenges, and basic practice stay free and useful without sign-in.

Source readiness work focuses on trustworthy intake: clear licenses, contribution docs, issue templates, safety checks, and no-vendor local setup.

When the repository is public, small corrections and verifiable student-facing improvements should be easier to review than broad rewrites.

Good first contribution lanes

Correct a concept definition, unit, graph label, worked example, or misconception note.

Improve keyboard access, contrast, touch targets, or screen-reader wording on a simulation surface.

Make learner-facing copy more direct: predict, change one variable, observe, explain, then check.

Add a focused regression test for a route, content schema, or simulation control.

Protected boundaries

Security reports, account support, billing support, and private user-data questions should not be handled in public issues.

Billing, auth, database migrations, ads policy, AI-cost features, license, brand, and governance changes need owner review first.

Official deployment secrets, real Wrangler config, real ads.txt, vendor dashboards, and private operator history stay outside the public source tree.

How to help now

Until public issue intake is opened, the safest way to help is to report confusing concepts, broken routes, accessibility problems, or concrete improvement ideas through the contact form.

Send feedback Read about the project