AMBER

Catch gray-market diversion. Recover the margin.

Built on Bright Dataverify_packet GREEN
net-of-tax gap
+EUR 10.75
per unit · signed, not VAT
recoverable margin
EUR 537,500
per year · buyer volume assumption
residential exits agree
6 exits
spread EUR 0.00
verify_packet
GREEN
chain of custody intact · exit 0
provenance scheme
ed25519
+ sha256 RFC 6962 Merkle
The verifier console · centerpiece

A signed evidence packet anyone re-verifies offline

Edit one byte and the verdict flips. The RED/GREEN below is the real verify_packet exit code, captured verbatim, not a hardcoded UI state.
interactive · break it, then fix it
VERIFIED
Chain of custody intact. Every capture hash, the Merkle root, and the ed25519 signature re-check under the signer's out-of-band public key. exit 0
working copy · facts.json (a Merkle leaf)edit, then re-verify
real verifier output (verbatim)exit 0
$ python -m amber.cli samples/live_packet --pubkey f2de2b5f14785372ced46288f3009448db17495312fe0492377fd14b036a5dc8
trusted signer source: --pubkey (CLI) (1 key)
verify_packet: samples\live_packet
  [OK  ] be-01: body sha256 ok (d330b0359d0842ec...)
  [OK  ] be-02: body sha256 ok (28b3183b444f0e26...)
  [OK  ] be-03: body sha256 ok (c53ce8fed489b745...)
  [OK  ] de-01: body sha256 ok (1127eb8160e177f9...)
  [OK  ] de-02: body sha256 ok (399a7dc4a44590f5...)
  [OK  ] de-03: body sha256 ok (1127eb8160e177f9...)
  [OK  ] merkle.json: leaf table matches recomputed leaves
  [OK  ] merkle.json/root: root ok (c5a6fc3887dfaf46...)
  [OK  ] signature.json: algorithm/scheme pinned: ed25519 + sha256 rfc6962
  [OK  ] signature.json: ed25519 signature verified over root under trusted signer f2de2b5f14785372...

  [OK] VERIFIED -- chain of custody intact
OK be-01: body sha256 ok (d330b0359d0842ec...)
OK be-02: body sha256 ok (28b3183b444f0e26...)
OK be-03: body sha256 ok (c53ce8fed489b745...)
OK de-01: body sha256 ok (1127eb8160e177f9...)
OK de-02: body sha256 ok (399a7dc4a44590f5...)
OK de-03: body sha256 ok (1127eb8160e177f9...)
OK merkle.json: leaf table matches recomputed leaves
OK merkle.json/root: root ok (c5a6fc3887dfaf46...)
OK signature.json: algorithm/scheme pinned: ed25519 + sha256 rfc6962
OK signature.json: ed25519 signature verified over root under trusted signer f2de2b5f14785372...
out-of-band trust

The verifier is pinned with --pubkey to the signer's independently-published public key f2de2b5f14785372ced46288f3009448db17495312fe0492377fd14b036a5dc8, not the repo's bundled allowlist. Edit the packet (or even its allowlist) and you still cannot forge a GREEN: the judge holds the key.

The catch · same SKU, two markets

Apple AirPods 4 (ANC, charging case) MXP93 GTIN 0195949689673

GTIN 00195949689673 (GTIN_MATCH) · dispatched same-second · a brand monitoring its OWN SKUs
🇩🇪
Germany
DE
VAT 19%
gross 179 EUR incl. VAT
EUR150.42
net of tax · purchasable · in stock
https://www.mediamarkt.de/de/product/_apple-airpods-4-mit-aktiver-gerauschunterdruckung-in-ear-kopfhorer-bluetooth-weiss-2954282.html
net-of-tax gap
+EUR 10.75
/ unit
signed, not VAT
🇧🇪
Belgium
BE
VAT 21%
gross 169 EUR incl. VAT
EUR139.67
net of tax · purchasable · in stock
https://www.mediamarkt.be/nl/product/_apple-draadloze-oordopjes-airpods-4-actieve-ruisonderdrukking-oplaadcase-mxp93zma-2152461.html
PRICE DELTA DETECTED · signed, net-of-tax, chain of custody intact

The signed packet records this fact only. A human draws any legal conclusion. Geo is EXIT_ONLY (the country is the proxy exit, not a GPS witness); captures were dispatched the same instant, not witnessed the same instant.

why it is not the tax

Germany has the lower VAT (19%) yet the higher net price, so stripping tax widens the gap (gross EUR 10.00 to net EUR 10.75). The gap is the channel, not the tax.

The business number

Recoverable margin per year

The signed per-unit delta, scaled by the brand's own diverted-volume estimate.
EUR 10.75
signed delta / unit
x
50,000
units / yr (assumption)
=
recoverable margin / year
EUR 537,500
buyer-supplied assumption50,000 units / yr is a buyer-supplied volume assumption, not an Amber measurement. Amber signs the per-unit delta; the brand supplies the volume.
The within-country control

Multiple residential exits per country, all in agreement

Many residential IPs per country, all dispatched the same second.
🇧🇪
Belgium
3 residential exits
intra spread
EUR 0.00
be-01agree
net of tax
EUR 139.67
exit 109.130.251.198HTTP 200
be-02agree
net of tax
EUR 139.67
exit 178.118.147.238HTTP 200
be-03agree
net of tax
EUR 139.67
exit 94.110.5.137HTTP 200
🇩🇪
Germany
3 residential exits
intra spread
EUR 0.00
de-01agree
net of tax
EUR 150.42
exit 77.22.56.14HTTP 200
de-02agree
net of tax
EUR 150.42
exit 62.155.220.38HTTP 200
de-03agree
net of tax
EUR 150.42
exit 93.207.216.185HTTP 200
All in-country exits agree to the cent (spread EUR 0.00), so the only thing we changed was the country: a controlled experiment, not exit-IP noise.