Public evidence

engineering.measurement.accuracy

Machine-checked verification for calculation version 1.0.1 · tier engine_tested

Open calculator API Capability Download JSON CVP Manifest Reproduce MCP Review policy

CVP status

Not Type A. O3 tabulated vectors only.

Declared partition coverage: PASS · 7/7 declared partitions (alias, awkward, ddof0, low, zero, high, invalid-domain)
Versions: Calculation 1.0.1 · CVP protocol 1.0.0-proposed · Evidence 2026-09-25.o2-o3
Verification revision: 2026-09-25.o2-o3 · 1/1 property · digest cc85d9119311

Verification

Legacy regression suite: 9/9 calculation tests · Production surface contract 6/6

Three distinct suites: educational examples on the calculator page, this legacy regression count (9/9 golden + boundary), and the CVP verification suite below. They are not the same number. Property, metamorphic, round-trip, O3, and interface statuses/checks are recorded in the CVP Manifest . SKIP (0/0) means that suite has no published vectors.

This calculator CURRENT · Public schema 1.0.1 matches · Semantic contract ✓ · Production attested · Public/cache ✓ · Origin ✓

CVP verification suite

Protocol checks from the Evidence Manifest. Distinct from the legacy regression suite above. Verification revision 2026-09-25.o2-o3. Calculation version 1.0.1. CVP 1.0.0-proposed. 4/4 golden · 4/4 CVP boundary · 4/4 invalid · 1/1 property · 1/1 metamorphic · 8/8 O3 · 2/2 cross-interface · 6/6 CVP contract. digest cc85d9119311. O2: signed bias vs a separate-module sample identity (≤2 ULP). Not Type A u_A and not a single-shot error. ≤2 ULP vs O3 applies only to the published tabulated bias vectors. Not Type A uncertainty.

Golden (oracle-backed) 4 / 4
CVP boundary 4 / 4
Invalid input 4 / 4
Property 1 / 1
Metamorphic 1 / 1
O3 high-precision 8 / 8
Cross-interface 2 / 2
CVP contract 6 / 6
Interfaces PASS · UI (SSR) / REST / MCP
Interface claim ui-ssr is query-result HTML, not a live browser session. Error-path engine·REST·MCP 1/1 (status, code, calculation_version). SSR compared on URL-canonical requested calculations; empty query is idle (not an error) and JSON-typed object/array inputs are REST/MCP-only.
Verified scope Signed bias x̄ − xm from repeats on a master, + O3 mpmath tabulated bias.
Assurance Engineering
Deferred Not Type A uncertainty and not a single-shot measurement error.

release artifacts aligned PASS
schema output required declared PASS
schema valid input executes PASS
engine success output validates schema PASS
schema rejects incomplete input PASS
rest success payload validates schema PASS

Fingerprint sha256:08a13dedfe6cb9ac87a4bef40979d0a3e1e7a388a685982912e53a3c5cd50531

Input schema sha256:dd207a5a0ffc822fb75b18576a250ff74d79e654909808201ac7633f084372d0 · Output schema sha256:e4b7a26f0681937e954120961cc1088c43205bcf402a55178378e7e9a113b5a8 · Build-time SHA-256 digests prove published artifacts agree with each other (release_integrity.scope=workspace). Production attestation GETs canonical URLs twice (public/cached view and origin), including Capability HTML, via npm run attest:production --write (published at /.well-known/calculatorx-production-attestation.json) and fails if the public Capability page advertises a different calculation_version. This is artifact consistency, not an independent immutable release log.

Build & scope

Engine identity and declared limitations.

Tool: accuracy
Engine: accuracy
Engine build: 2026-09-24
Evidence core: 2026-09-24T22:06:06.994Z
Rendered: 2026-09-30T13:48:16.163Z
Last reviewed: —

Assumptions

VIM trueness: signed bias = x̄ − xm from repeats on a master. s may be zero.
Not a single-shot measurement error, not Type A u_A = s/√n, and not Type-1 Cg (no T).

Limitations

Signed bias from a sample mean — not a single-shot measurement error ε
Does not report Type A u_A = s/√n
Does not need a tolerance T (that is Type-1 Cg/Cgk)
s = 0 is valid; n = 1 with ddof = 1 is VALUE_OUT_OF_RANGE

Sources

Standards and references supporting this tool specification. Methods are listed separately — they are algorithms, not bibliographic sources.

JCGM 200:2012 — International vocabulary of metrology (VIM) — 2.13 measurement accuracy; 2.14 measurement trueness; 2.18 measurement bias
Supports: Accuracy is qualitative; trueness is closeness of the mean to a reference; bias is the estimate of a systematic measurement error
ISO 5725-1 — Accuracy (trueness and precision) of measurement methods and results — Trueness as bias of the mean versus a reference
Supports: This page reports signed bias of the sample mean; it does not report a precision index as the product

Legacy published checks

Golden and boundary cases from the legacy Evidence JSON. CVP property, metamorphic, round-trip, O3, and interface statuses/checks are recorded in the Manifest — including SKIP when a suite has no vectors.

ID	Kind	Status	Detail
`golden-acc`	golden	pass	Expected bias=-1 · abs_bias=1 · xbar=7 · s=1 · Actual n=3 · bias=-1 · abs_bias=1 · xbar=7 · s=1 · xm=8 · ddof=1
`boundary-acc-zero`	boundary	pass	Expected bias=0 · s=1 · Actual n=3 · bias=0 · abs_bias=0 · xbar=8 · s=1 · xm=8 · ddof=1
`boundary-acc-flat`	boundary	pass	Expected bias=0 · s=0 · Actual n=3 · bias=0 · abs_bias=0 · xbar=8 · s=0 · xm=8 · ddof=1
`boundary-acc-n1`	boundary	pass	Expected VALUE_OUT_OF_RANGE · Actual VALUE_OUT_OF_RANGE
`boundary-acc-missing`	boundary	pass	Expected MISSING_REQUIRED_INPUT · Actual MISSING_REQUIRED_INPUT
`property-acc`	property	pass	-1
`property-acc-zero`	property	pass	0
`property-acc-flat`	property	pass	0
`property-acc-alias`	property	pass	-1