Most governance tools ask you to trust a dashboard. This one you can verify, run, or buy.

VasperaMesh wasn’t designed in a slide deck and then built. It was extracted from a system that already governs itself in production. The same deterministic core appears at three rungs — proof, open core, product — and each one is checkable against the one below it.

ryancolkitt.com publishes autonomously from public data. Nothing goes out unless a deterministic gate passes it: sources present, figures grounded, disclosures attached. Every decision — held or published — is written as one immutable, hash-chained row, and the chain head is anchored on Bitcoin. It is tenant one of this product, running the exact code you can buy.

You don’t have to take that on faith. The verifier re-derives every hash from the published spec and cross-checks the anchor, importing nothing from us.

The decision spine of that proof — the declarative gate, the RFC 8785 canonical hashing, the chain math, and the standalone verifier — is published as @vaspera-capital/mesh-core. It is dependency-light and stack-agnostic: the same package the proof imports, golden-vector tested, with no vendor in the decision path. If you want to own the code that makes and records your decisions, you can run just this.

A regulated lender doesn’t want a library; they want the core operated for them and defensible to an examiner. VasperaMesh wraps mesh-core with the things that turn a decision engine into a control plane: one-line OpenTelemetry capture, a plain-English policy lifecycle with separation of duties, inline enforcement (shadow → advisory → enforcing), multi-tenant isolation, framework-mapped evidence packs, and — for credit — ECOA reason codes and §1002.9 adverse-action notices. Run it as managed SaaS or self-hosted in your own VPC; either way the record is the same one anyone can verify offline.