An official website of the Disclosure Foundation
1RecordsThe unit of record2ObservablesDetection3OriginsClassification4ClaimsAttribution5ScoringEvaluation6InstrumentsThe instrument layer
DocsOnboardChangelog

The Standard

1RecordsThe unit of record2ObservablesDetection3OriginsClassification4ClaimsAttribution5ScoringEvaluation6InstrumentsThe instrument layer

Docs

Implement the standard

Onboard

Bring your dataset

Changelog

What changed, and when

GitHub

Open standards and tools for UAP research.

The Standard

  • Records
  • Observables
  • Origins
  • Claims
  • Scoring
  • Instruments

Resources

  • Documentation
  • Onboard your data
  • Changelog

Project

  • GitHub
  • Disclosure Foundation
  • Donate

© 2026 Disclosure Foundation. All rights reserved.

The Standard · Part 6 of 6 — The instrument layer

Instruments

Organizations publish what their sensors are, how they keep time, and how they are calibrated — so data from different arrays can be compared honestly, and well-documented instruments earn the trust they deserve.

@disclosureos/instrumentsDocumentation

A published catalog, not a per-case form

Everything so far in the standard lives inside one case record. The sensor manifest is different: it is the standard’s first standalone document. An organization that operates detection hardware — a research group, a vendor, an institution — publishes one manifest describing every sensor it runs. Case records then point at manifest entries from their sensor readings, so the full hardware story is written once and cited everywhere.

Community contribution

The sensor manifest was proposed by ELDÆON, the standard’s first external contributor, and promoted into the framework as @disclosureos/instruments. Their published manifest is the worked example below.

  • schemaVersionstringrequired

    Version of the sensor-manifest schema this instance targets (SemVer).

  • orgstringrequired

    Publishing organization, display form (e.g. "ELDÆON").

  • orgSlugstringrequired

    Stable lowercase slug for the publisher (e.g. "eldaeon"); the org half of a `sensorRef`.

  • orgUrlstring

    URL for the publishing organization.

  • generatedAtstring

    ISO date the manifest was generated.

  • sensorsSensorEntry[]required

    One sensor: its class, spec, timing, raw-data locator, measurements, and calibration provenance.

  • futureUpgradesFutureUpgrade[]

    Planned hardware declared ahead of deployment so data consumers can anticipate new streams. Not part of the live sensor array.

What each sensor declares

A case record says a radar tracked something. A manifest entry says what that radar is: the kind of instrument, the clock that timestamps its data, the units and error bars of what it measures, where its raw files live, and how — or whether — it is calibrated.

Sensors are organized across 6 modalities: Audio, Electro-Optical, Radio Frequency, Quantum, Environmental, Biometric.

Records mapping

recordsMapping
sensorType
string
detectionMethod
string
proposedSensorType
boolean
proposedDetectionMethod
boolean

Timing

timing
timeSource
enum
timeUncertaintyNs
unknown

Raw data

rawData
format
string
locatorPattern
unknown
sampleRateHz
unknown
bitDepth
unknown

Measurements

measurements
name
string
unit
string
uncertainty
MeasurementUncertainty

Calibration

calibration
status
enum
calibrated
unknown
currentMethod
unknown
traceableReference
unknown
referenceInUse
unknown
cadence
unknown
uncertaintyBudget
unknown
notes
unknown

Time sources

A timestamp is only as good as the clock behind it. Timing error of even a few seconds confounds any attempt to correlate two stations’ data, so the clock source is declared, never assumed.

  • Atomic Clockatomic_clock
  • GPS-Disciplinedgps_disciplined
  • NTPntp
  • System Clocksystem
  • Unknownunknown

Calibration is a gradient, not a checkbox

Most datasets reduce calibration to a yes/no box. The manifest records it as a maturity gradient instead, and “no calibration is performed” is a valid declared state rather than a blank. Stating where each instrument sits makes calibration comparable across organizations and gives every operator a defined path from one rung to the next.

  • 01
    Nonenone

    No calibration is performed. Declared honestly, this is itself useful information.

    ×1.00
  • 02
    Candidate Identifiedcandidate_identified

    A traceable reference has been named but is not yet part of practice.

    ×1.05
  • 03
    In Practicein_practice

    A real calibration method is performed today, but no full uncertainty budget exists.

    ×1.15
  • 04
    Documenteddocumented

    The calibration method and its uncertainty budget are published.

    ×1.25

The multipliers on the right are where this feeds back into scoring: a claim backed by an instrument with published calibration provenance earns additional weight in the consensus score. The defaults are deliberately conservative — a sensor without a manifest entry keeps the ×1.00 baseline, and no rung of the gradient ever weighs less than that baseline. Publishing provenance can only add weight.

One manifest, published

Everything in this section comes straight from ELDÆON’s published manifest — 28 sensors across 6 modalities, plus 3 declared future upgrades. The page renders the document itself, so as the manifest evolves, this section follows.

Sensors by modality

  • Audio2
  • Electro-Optical6
  • Radio Frequency3
  • Quantum2
  • Environmental10
  • Biometric5

Calibration status across the array

  • None5 sensors
  • Candidate Identified13 sensors
  • In Practice9 sensors
  • Documented1 sensor

Most of the array sits mid-gradient, working toward documented calibration, and the manifest states exactly where each instrument stands — which is precisely what the format is designed to capture.

Infrasonic Microphone

Candidate Identified

eldaeon:dionysus-infrasonic-mic

Hardware
Barometric differential-pressure sensor
Time source
GPS-Disciplined
Raw data
NDJSON.gz

Passive Radar (bistatic SDR)

In Practice

eldaeon:dionysus-passive-radar

Hardware
2-channel coherent SDR receiver (reference + surveillance antennas)
Time source
GPS-Disciplined
Raw data
JSON (delay-Doppler maps, detections, tracks) + bounded raw IQ capture
Calibration
continuous per-detection truth association against ADS-B aircraft converted to bistatic delay-Doppler coordinates (open-source adsb2dd); match tolerance 2 km delay / 5 Hz Doppler with residuals recorded

Fluxgate Magnetometer (3-axis)

Candidate Identified

eldaeon:dionysus-fluxgate-magnetometer

Hardware
3-axis fluxgate magnetometer
Time source
GPS-Disciplined
Raw data
NDJSON.gz

Published manifests

Each organization hosts its own manifest; a small shared registry maps organization slugs to published manifest locations, so a sensorRef like eldaeon:dionysus-passive-radar can always be resolved. Adding your organization is a pull request.

  • ELDÆONeldaeonView manifest

Operate sensors? Publish a manifest and validate it with disclosureos manifest validate.

Part 5 of 6ScoringEvaluationReady to buildDocumentationImplement the standard