What is live
No figures that go stale. The capability catalogue lives elsewhere; this is the maturity an investor needs so as not to buy a shadow.
| Layer | Status |
|---|---|
| On-prem product: install, institute, commission, deliver with a seal | [V][Verified] in operation, closed beta |
| Persistent agent identity, private corpus, Task Queue, bilingual Console | [V][Verified] |
| Authentication (OIDC) and roles on the backend, not just on buttons | [V][Verified] |
| Governance as a tray: approve means apply | [V][Verified] built; generic enforcement is off by default [P][Partial] |
| Selection ladder (fitness): lower and middle rungs | [V][Verified] measured; the reliability gate already converges in development but is not closed: its stability on new data and a measurement on a client installation are still missing [P][Partial]; routing rungs in shadow, switched off by decision [P][Partial]; recursive self-improvement not started [D][Designed] |
| Ontology catalog (RDF / JSON-LD, closed vocabulary, SHACL) | [V][Verified] deployed as a surface distinct from the live graph; [P][Partial] no in-catalog tenancy, no automatic sync |
| Host federation (audited envelopes, trust policies) | [V][Verified] transport; [P][Partial] envelopes are not cryptographically signed |
| SaaS multi-tenancy | [D][Designed] last by decision |
Self-improving, precisely. Variation (the agent changes skills and memory) already exists in the runtime. Measured selection is the ladder: today it recommends; it does not act on live routing, and that is a decision, not the gate holding it back. Saying “the system improves itself” skips a rung that is not there.
Commitment: we do not sell decorative governance or standard-name conformity without evidence in code. See Governance and Limits.