math.percent_change

Three distinct suites: educational examples on the calculator page, this legacy regression count (8/8 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.

STALE · Build schema 1.1.1 matches page · production binding STALE · Semantic contract ✓ · Last attestation PASS · current evidence changed · re-attestation required · CVP STALE · attestation STALE — Production CURRENT withheld · Public/cache ✓ · Origin ✓

CVP verification suite

Protocol checks from the Evidence Manifest. Distinct from the legacy regression suite above. Verification revision 2026-10-22.percent-change-complete-contract-v3. Calculation version 1.1.1. CVP 1.0.0-proposed. 108/108 golden · 2/2 CVP boundary · 25/25 invalid · 5/5 property · 1/1 metamorphic · 4/4 round-trip · 7/7 cross-interface · 3/3 cross-calculator · 6/6 CVP contract. digest c76f6f160ef8.

• Golden (oracle-backed) 108 / 108
• CVP boundary 2 / 2
• Invalid input 25 / 25
• Property 5 / 5
• Metamorphic 1 / 1
• Round-trip 4 / 4
• Cross-interface 7 / 7
• Cross-calculator 3 / 3
• 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 2/2 (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.
• Assurance Core

• 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 sha256:554dc7588111216564ee2b53e5d72d2798bcb0124ebd3fb6a797ade2c5a9f165

Input schema sha256:f79b06c286ee7e4fc44ee2ffc438f5f196fe0916dd8500e05340a65c5b1aef9a · Output schema sha256:3bfdc00173d02aaf9da8bd155d8633952ed38f7d7a67f2a9fa18fca67ac8ef7c · 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.

Tool

percent-change

Engine

percent-change

Engine build

2026-10-22

Evidence core

2026-10-22T01:11:44.702Z

Rendered

2026-10-22T08:54:35.069Z

Last reviewed

—

Assumptions

• Interactive evaluation runs locally in the browser. Shared URLs and REST evaluate the same versioned engine (query SSR).
• CVP interface evidence is UI (SSR) query-result HTML, REST, and MCP (MCP reuses the calc API handler). That is interface-result consistency, not a live browser session, cache switching, or end-to-end UI test.
• Signed-change convention: denominator is |V₁|, so a move from −100 to −50 is a +50% increase. Some definitions use V₁ without absolute value or restrict to positive baselines.
• Two active fields normally determine the output. Explicit solve_for selects the output. Choose Check all3 to enter all three and check consistency. Shared URLs and machine inputs with all three also check consistency before solving.
• Solving for V₁ is not always unique. The engine returns both valid initials when they exist.

Limitations

• Signed change uses |v1|; nonzero initial value required. Inverse preserves both signed branches or infinite/no-solution outcomes.
• Exact rational expressions of normalized binary64 inputs, one ties-even rounding per published field; not exact decimal measurement arithmetic.
• Reject nonzero underflow/overflow in inputs, secondary fields or valid candidate branches; no partial candidate set. Rescale. Near±100 inverse percentages are ill-conditioned.
• 8ULP same-sign nonzero triple consistency, strict zero; preserve requested percent when solving final even if final rounds unchanged.