This is one researcher and an open repo. It needs you.

Both papers are preprints seeking review. Tell us where the theory is overstated, where the engineering choices are wrong, where the citations don't hold. Adversarial readings are welcome; the evaluation framework was built to be proven wrong.

The reference implementation runs on a consumer workstation: 64 GB RAM, one or two commodity GPUs, fully offline if you want. Replicate the seven experiments from their seeds, and report what you find. Null results are reportable results.

CAL v0.2 is legally novel and untested: AGPL backbone, neurorights-derived covenants, copyfarleft commercial terms, a guardianship pathway. We need counsel willing to review it and tell us where it breaks before someone else finds out in court.

Commit code to the reference implementation. The future-work list is long: a learned workspace, V-JEPA 2 for the world model, speaker diarization, a symbolic reasoning module, containerized deployment, Fieldtrip, Paracosmic, and the volunteer-compute substrate.