Cross-organization authorization
Share issuers, status and authorization decisions across institutions while personal claims and internal evidence stay inside the organization that holds them.
Qkite is a permissioned blockchain for applications that span several organizations. Everyone shares one ordered, auditable history. Nobody has to hand their confidential records to a central operator — or to every validator that verifies the chain.
Wasm contracts run through a narrow, resource-bounded host interface. Same inputs, same result, on every machine that runs them.
ML-DSA signatures and ML-KEM key establishment, with standard symmetric encryption and key derivation protecting requests and stored data.
Who may hold a collection and who must attest to a function are committed to the ledger. The values behind them stay encrypted.
Either a central operator becomes the source of truth, or every participant receives the data needed to reproduce every operation. Neither works when independent organizations must coordinate while keeping control of confidential records. Switch between the three models to see what actually changes.
A single operator becomes the source of truth. Coordination is easy, and the compromise is structural: one party can read, correlate, withhold or lose everything, and the other participants have no independent way to check the history.
Classic replication earns independent verification honestly — by giving every participant the inputs required to recompute every operation. That is exactly the wrong property when the inputs are prices, balances, personal claims or regulated records.
The public ledger records organizations, keys, contract code, policies, public application state, and opaque continuity commitments for private collections. Private values stay encrypted in stores run by the collection's members. Validators enforce the exact committed policy and verify signed results — without receiving the private payload or replaying private execution.
Confidentiality in Qkite is not a promise about intent — it is a property of what each participant is handed. Chaincode still controls what it deliberately derives into public output, so confidentiality also depends on correct application logic.
In the current prototype the finalizer is trusted for ordering and availability — never with private plaintext. Public replicas verify the complete public history on their own, with no private database and no private Wasm replay.
Six steps, deliberately narrow. Select a step to light up the part of the system it touches.
Collection access says who may hold the data. Function attesters say who must agree on a result. Change the shape below and watch the Go chaincode — and the failure mode — change with it.
Qkite is horizontal infrastructure — not a claim that every database should become a blockchain. Its value shows up when several organizations need a common deterministic result, independent verification, and selective disclosure at the same time.
Share issuers, status and authorization decisions across institutions while personal claims and internal evidence stay inside the organization that holds them.
Agree on obligations and final outcomes while commercial terms, balances and settlement instructions remain restricted to the counterparties.
Share custody and provenance across a supply chain while prices, quantities and organization-specific operational data stay private.
Coordinate auditable decisions among institutions while regulated records stay within the organizations authorized to hold them.
The public-ledger and deterministic-contract foundation is established. Same-machine feasibility evidence exists for the confidential executor design — independent organizations executing and attesting to private operations without disclosing private inputs to public replicas. That evidence is not counted as a shipped runtime.
A bounded four-process prototype has already exercised the confidential path end to end:
0 of 8 complete — the minimum path from the current state to the initial usable version. Open an item for its completion condition.
Completing these defines the initial usable version: an operable multi-host, multi-organization permissioned blockchain with distributed finality, confidential execution, packaged node and CLI operation, and a usable chaincode environment.
Later work will be driven by actual use rather than a speculative long-range plan. The currently visible areas:
The reported baseline is the public ledger and the deterministic Wasm contract environment. The executor runtime stays a todo item while its experimental same-machine implementation is stabilized and formally accepted. That prototype uses fresh databases, one finalizer, explicit static loopback peers and same-machine process isolation. It does not yet claim:
Exact already-finalized submissions are deduplicated in the prototype, but an in-flight ambiguous request is not yet automatically recovered or resubmitted. We would rather publish that list than a launch date.
What has been demonstrated is the central result: independent organizations can share deterministic application outcomes and one auditable public history while authorized private data stays outside that public ledger. The work ahead extends experimentally validated protocol behavior — it does not start from an untested architecture.
There is no download, no SDK release and no public network yet. When the executor runtime is formally accepted, that changes first.