Technical brief

Model output is intent, not authority.

Aukora is a cryptographic authority and receipt layer for AI agents. A model may propose an action; a separate kernel decides whether it happens, scopes the permission to a single use, and records the outcome — allowed or refused — as a signed, hash-chained receipt anyone can verify. This brief is for people evaluating the system, so it states plainly what is working today and what is still designed.

The model proposes. The kernel decides. The receipt proves.

1 · Architecture

Aukora separates three planes that most agent stacks fuse together:

The kernel

A small policy layer that classifies every proposed effect and decides allow / refuse. It is deliberately not the model, not the editor, and not the tool — it is the checkpoint they all pass through.

Authority

Permission to cause a specific effect. It is scoped to that effect, consumed exactly once, and never held ambiently by the model. No standing or blanket grants.

The receipt chain

An append-only, hash-chained log where each entry names the one before it by hash and is signed. It is the trust object — not a dashboard, not the model's prose.

Because the boundary lives at the one line where any agent reaches for a tool, the host — an IDE, an API, an MCP server, a workflow, an agent runtime — stays swappable. Aukora attaches to the host's action surface rather than replacing it.

2 · Core invariants

These are the load-bearing rules. They hold regardless of which model, engine, or tool is in play.

  1. Model output is intent, not authority. A model can propose, draft, rehearse, and argue for an action. It can never be the thing that authorizes it.
  2. A separate kernel decides. The component that judges an action is not the component that produced it. Advice — even unanimous multi-model advice — is evidence, never a grant.
  3. Authority is scoped and consumed once. A permission authorizes one specific effect and cannot be replayed for another.
  4. Every protected effect leaves a receipt — and so does every refusal. A blocked action is recorded, not silent.
  5. High-risk effects require a human signature. They are bound to proof that a person approved the exact scope, made deliberately, every time — no silent self-authorization.
  6. Evidence is portable. The receipt chain can be verified by a second party without trusting the model's claims or the machine that produced them.

3 · The authority flow

End to end, a protected action moves through the same path:

intentclassify (ring / risk)gate: allow / refusescoped one-time tokenmediated effectsigned receiptindependent verify

Low-risk reads and drafts flow freely. The moment an action would change your world — write a file, call a tool, touch data, reach outside the machine — it is classified, gated, and (if allowed) executed only through a permission minted for that one effect. Refusals short-circuit before any effect and are themselves receipted.

Typical agent stack
modeltooleffect

Output becomes an action directly. No authority boundary, no record.

Aukora-wrapped
modelintentgatetokenmediated toolreceiptverifier

The effect happens only through a scoped, one-time token — and it's provable afterward.

4 · The receipt

A receipt is the atom of evidence. Each one records what was authorized and what happened (or was refused), names the receipt before it by hash, and is signed — so rewriting recorded history breaks the chain and shows. From that single object several proofs can be derived without trusting the model:

Authoritywho allowed this
Actionwhat actually happened
Refusalwhat was blocked
Provenancewhere a memory, datum, or output came from
Usagewhat billable action or compute occurred
Audita second party can re-verify the chain
Trainingoutcomes trustworthy enough to be a label
Attributionwhich model or agent actually produced it

Signatures are Ed25519. Content can be erased on request (right-to-be-forgotten) while the chain still proves the erasure occurred. The exact on-disk record format is being stabilized ahead of public release; treat field-level specifics as not-yet-final.

5 · What is working vs designed

Conservative on purpose. "Lab" means proven in the working tree with a large real test suite; "dev-real" means running but still hardening; "research lab" means running live in the experimental lane, not yet merged into the core; "designed" means specced, not yet built; "research · direction" means a direction being explored — nothing built yet.

LayerStatus
Authority kernel · scoped consumed authorityworking · lab
Signed, hash-chained receiptsworking · lab
Local governed coding loopworking · lab
Human approval gate (AUMLOK)dev-real · hardening
Cross-node verificationworking · expanding
Model-portable identity (substrate swap)working · lab
Time-anchored records (drand beacon)research lab · live
4D Gaussian memory fieldresearch · direction
Tool / MCP gatewaydesigned · expanding
Receipt-certified trainingdesigned · partial
Finance / enterprise adaptersfuture · partnership path
Public open-source releasesequenced with release plan

6 · Integrating a tool

Adoption is at the effect boundary, not a rewrite. Conceptually:

  1. Identify the effect you want governed (a shell call, an API request, a broker order).
  2. Route that call through the gate: the agent submits it as intent, not a direct call.
  3. The kernel classifies and decides. If allowed, it mints a one-time token scoped to that effect.
  4. The effect executes only with that token, then a signed receipt is written.
  5. Any party can later verify the receipt chain independently.

Aukora can also help generate adapters for a new host, but a generated adapter gains no authority automatically. New adapters must be approved, scoped, tested, and activated through the gate — the same rule as any other effect.

7 · Limitations — what it does not claim

See the loop Release status ↗ For partners & enterprises