pki.sgit.ai / documents / observability

Observability Is The Usage Graph Nobody Has To Declare: Check Events Belong In The Issuer's Own Lane

TypeArchitecture brief Versionv0.33.61 Date20 August 2026 AuthorDinis Cruz (project lead) and collaborators LicenceCC BY 4.0 Sourceraw markdown · view on GitHub

Summary

The observability layer, and the one decision inside it that separates a security property from a surveillance product. Observability goes in at the beginning, not because it is good practice but because the case for declared mandates is that they produce the evidence saying where enforcement earns its cost — and without check events they produce none. The correction the proposal needs: a verification is not a use, and the error runs both ways, since a party that uses a mandate without verifying generates nothing while a resolver walking the chain generates an event with no usage behind it. So the primary output is the missing edges — the parties holding a mandate that have never once checked it — which the issuer can compute because it holds both halves. Where the events are written decides what this is: a central log accumulates who is evaluating whom across parties that never consented; the issuer's own lane makes it an owner observing their own asset. And because nothing is pushed, the interval between a party's checks is its effective revocation latency, measurable before anything is ever revoked.

Key concepts

Key ideas

On this site

Became document 11 of the registry MVP pack, and answers a question the mandate page had been raising since it was published — by refusing it: nobody can say who is using a mandate, and the issuer can say who has never checked one.

Read the document

📄 Original document · v0.33.61 · 20 August 2026 · rendered from the raw markdown (the source of truth)