☰ Documentation menu
Libraries & reference

The compliance standards QuantumLedger ships with, generated live from the definitions currently loaded. Each control lists what the standard requires, the automated checks used as evidence, and how to remediate a gap.

FAIR Data Principles v1.0

Findable, Accessible, Interoperable, Reusable. Applied to quantum result records: every published result carries a persistent identifier, is retrievable, uses an open interoperable circuit format, and declares a licence.

FAIR-A1 — (Meta)data are retrievable by their identifier via a standard protocol (medium)
The result record must be retrievable — i.e. published as a ResultCard that is publicly accessible.
  • Each published result card must satisfy: card.published_at is present and not empty. FAIR-A1-published
Remediation: Publish the ResultCard (set visibility=public and a published_at).
FAIR-F1 — (Meta)data are assigned a globally unique and persistent identifier (high)
Every published result must be identified by a globally unique, persistent identifier (a DOI or other PID) so it can be found and cited unambiguously.
  • Each published result card must satisfy: card.doi or card.pid is present. FAIR-F1-pid
Remediation: Mint a DOI (or register a PID) for the published ResultCard and store it on the card before publishing.
FAIR-I1 — (Meta)data use a formal, accessible, shared, broadly applicable language (medium)
The circuit is stored in an open, interoperable representation — QASM3, the QuantumLedger IR (qlir/1.0), or QIR — not a vendor-locked binary.
  • Each circuit must satisfy: circuit.fmt is one of [qasm3, qlir/1.0, qir]. FAIR-I1-fmt
Remediation: Re-record the run capturing the logical circuit as qasm3, qlir/1.0 or qir.
FAIR-R1.1 — (Meta)data are released with a clear and accessible data usage licence (high)
Every published result must declare a licence so downstream users know the terms of reuse.
  • Each published result card must satisfy: card.license is present and not empty. FAIR-R1.1-license
Remediation: Set a licence (e.g. CC-BY-4.0) on the ResultCard before publishing.

IEEE P7131 Quantum Computing Reproducibility v2025

Reproducibility of quantum computations: recorded execution environment, time-bounded device calibration, tamper-evident run hashing, and at least one verified reproduction of a result in the workspace.

P7131-CAL — Device calibration snapshot is bound and fresh (high)
Each run binds a calibration snapshot captured within 30 days of the run so device state is reproducible; stale calibration is flagged as drift.
  • Each run must satisfy: run.calibration_snapshot_id is present. Evidence must be no older than ~30 day(s) (relative to run.finished_at). P7131-CAL-bound
Remediation: Re-capture a current calibration snapshot and re-run.
P7131-ENV — Execution environment is recorded (high)
Each run records the backend it executed on and the transpiler seed, so the execution environment can be reconstructed deterministically.
  • Each run must satisfy: all of run.backend_id, run.seed_transpiler are present. P7131-ENV-present
Remediation: Re-record the run through the SDK so backend_id and seed_transpiler are captured.
P7131-HASH — Runs are cryptographically hashed (high)
Each run carries a Merkle-bound run_hash so any tampering with its inputs is detectable.
  • Each run must satisfy: run.run_hash is present and not empty. P7131-HASH-present
Remediation: Seal the run so its run_hash is computed and chained.
P7131-REPRO — At least one verified reproduction exists (high)
The workspace contains at least one independently verified reproduction event, demonstrating that a recorded result can be reproduced.
  • Each workspace must satisfy: the number of workspace.reproduction_events where reproduction.status equals 'verified' is >= 1. P7131-REPRO-count
Remediation: Reproduce a stored run and have the reproduction verified.

Quantum Metrology Traceability v1.0

Measurement traceability for quantum results: every reported result is bound to a recorded calibration snapshot of the measuring device and declares its shot count (measurement sample size).

MET-INTEGRITY — Reported counts are content-addressed (medium)
Each result carries a SHA-256 digest of its counts so the reported measurement cannot be silently altered.
  • Each measurement result must satisfy: result.counts_sha256 is present and not empty. MET-INTEGRITY-hash
Remediation: Re-seal the result so its counts_sha256 is computed.
MET-SHOTS — Measurement sample size is reported (medium)
Each result reports the number of shots (measurement repetitions) so the statistical uncertainty of the reported distribution is quantifiable.
  • Each measurement result must satisfy: result.shots is present and not empty. MET-SHOTS-present
Remediation: Re-record the run so the result shot count is captured.
MET-TRACE — Measurement is traceable to a calibrated device state (high)
Each run binds a calibration snapshot recorded within 90 days of the run, so the measurement is traceable to a known instrument state.
  • Each run must satisfy: run.calibration_snapshot_id is present. Evidence must be no older than ~90 day(s) (relative to run.finished_at). MET-TRACE-calib
Remediation: Bind a current calibration snapshot to the run.

Reproducibility Policy v1.0

Organisation-internal policy: runs are deterministically seeded and cryptographically hashed, and the workspace demonstrates verified reproduction.

REPRO-HASH — Runs are hash-bound to a workspace chain (high)
Each run carries a run_hash chaining it into the workspace's append-only ledger.
  • Each run must satisfy: run.run_hash is present and not empty. REPRO-HASH-present
Remediation: Seal the run to compute and chain its run_hash.
REPRO-SEED — Runs are deterministically seeded (medium)
Each run records a transpiler seed so the compilation is deterministic and reproducible.
  • Each run must satisfy: run.seed_transpiler is present and not empty. REPRO-SEED-present
Remediation: Re-record the run supplying a transpiler seed.
REPRO-VERIFIED — Workspace has a verified reproduction (high)
At least one reproduction in the workspace has been verified before results are trusted.
  • Each workspace must satisfy: the number of workspace.reproduction_events where reproduction.status equals 'verified' is >= 1. REPRO-VERIFIED-count
Remediation: Reproduce a run and verify the reproduction.