EC EpochCore
PLATFORM ATLAS · QUANTUM WATERMARKING · STRATEGIC FISHBOWL

The platform is a provenance engine, not a watermarker.

What started as a single video watermarker is now a four-tool, in-browser media-provenance suite: mark, seal, scan, and verify — every byte processed locally, anchored to post-quantum signatures, distributed as one self-contained file. This atlas maps what exists, proposes an enterprise structure, and names what's missing.

build wm-3ae92a09d0cf version 2.0.0-universal-seal ras 40668c787c463ca5 coherence 0.9978 surface 100% in-browser
01 · THE FISHBOWL

Five layers, one substrate.

Read top-down: what the user touches, the engine beneath it, the cryptographic provenance it anchors to, the brand it wears, and how it reaches the world. Each chip is a real artifact in the tree — colored by whether it ships today.

Layer 5Surface · Tools
Watermarker Universal Sealer Merge Suite ×5 Verifier Launcher / front door Recipient key manager
Layer 4Engine · Capability
invisible-mark.js file-scanner.js · 9 daemons universal-sealer.js mp4-muxer · webcodecs kem-envelope-client.js audio-mark.js Tier B re-encode survival
Layer 3Provenance · Anchor
QCCF fingerprint QRNG entropy three-sig · ed25519+ml-dsa+slh-dsa WORM ledger D-KaP seal ML-KEM-1024 envelope
Layer 2Brand · Design System
colors_and_type.css ui_kits · app + marketing brand-kit · sigils preview ×29 cards tools off-brand (gold, not violet) Geist wordmark font
Layer 1Distribution
portable.html · 367KB self-contained flash-sync · WATERSEAL.json epoch-core-worker · CF apex shared lib / vendor (DRY)
ships today partial / off-brand / blocked missing
02 · CURRENT INVENTORY

What's on disk today.

The honest count, grouped by domain. Engine modules are proven (headless BER/PSNR, scanner attack corpus, real-browser export). The duplication and the scattered tools are the structural debt.

Watermarker engine 11 modules · proven

epochcore-watermarker/
  • invisible-mark.js — spread-spectrum luma markBER 0 · 45dB
  • watermark.js — visible logo + encode orchestrationH.264
  • file-scanner.js — 9-daemon vuln swarm7/7 attacks
  • universal-sealer.js — manifest + .epochsealAES-256
  • seal-ui.js · stack-panel.jsUI
  • mp4-muxer.min.js · webcodecs-export.jsmedia
  • kem-envelope-client.js — ML-KEM-1024needs bundle
  • epochcore-api.js · receipt-drawer.jsprovenance

Merge suite 5 tools · standalone

merge/ · merge2/
  • Combine Videos.htmlmerge
  • Volume Splice.htmlaudio
  • One-Word Overlay.htmlcaption
  • Fix Pronunciation.htmlaudio
  • Auto-Merge Two.htmlseam
  • off-brand palette · not sealed · dupes api.jsdebt

Design system canonical

colors_and_type.css · ui_kits/ · assets/
  • colors_and_type.css — two faces, one substratetokens
  • ui_kits/epochcore-app/ — App face8 jsx
  • ui_kits/epochpay-marketing/ — specimen11 jsx
  • _shared/ — brand-mark, seed-lattice, palettesigils
  • preview/ — design-system cards29
  • Geist wordmark font — system-ui fallbackcaveat

Brand & assets complete

assets/
  • brand-kit/ — logo, icon, swarm sigil (SVG+PNG)7
  • product-packs/ — 5 SMB coverspng
  • logos/ — IBM Business Partner lockups2
  • animations/ — Atoms, Scenes, Runwayjsx
  • screenshots/ — proof captures4

Distribution live

flash-sync/ · epoch-core-worker/
  • portable.html — self-contained, offline367KB
  • WATERSEAL.json — 12-asset hash manifestsealed
  • HANDOFF.md — propagation runbookdoc
  • epoch-core-worker — CF apex serve+verifygated
  • PQC_CRYPTO_BUNDLE_SPEC.md — awaiting buildspec

Provenance dataplane remote · live

qpc.q-routed.com · pqc.epochcoreqcs.com
  • QCCF / QRNG / qverify — three-sig200
  • WORM ledger — propose / applyanchor
  • ML-KEM-1024 registry — recipient wrapFIPS-203
  • browser ↔ dataplane — hash-only by designPHI-safe
03 · PROPOSED ENTERPRISE STRUCTURE

One monorepo. Apps, shared packages, design system, tests.

The fix for the structural debt: tools become apps/, engine modules become versioned packages/ (kills the duplication), the design system gets its own root, and the ephemeral proofs become a permanent tests/ harness. Color-coded by what exists vs. what to create.

epochcore-platform/ · proposed layout ◆ have ◆ partial ◆ create
epochcore-platform/
├─ README.md  # have
├─ VERSION.json  # create — single source of build/version truth
├─ PLATFORM_ATLAS.html  # this document

├─ apps/  # user-facing tools — each self-contained
│  ├─ watermarker/  # have — video mark + seal
│  ├─ sealer/  # partial — currently fused into watermarker; split out
│  ├─ verifier/  # CREATE — drop file+seal → verdict; drop video → recover mark
│  ├─ merge/  # have — combine · splice · overlay · pronounce · auto-merge
│  └─ launcher/  # CREATE — single front door routing to every tool

├─ packages/  # shared, versioned libraries — kills the duplication
│  ├─ crypto/
│  │  ├─ invisible-mark.js  # have
│  │  ├─ kem-envelope-client.js  # partial
│  │  ├─ epochcore-pqc-crypto.js  # CREATE — vetted noble bundle (local esbuild)
│  │  └─ audio-mark.js  # CREATE — spread-spectrum for audio-only files
│  ├─ provenance/
│  │  ├─ epochcore-api.js  # have — QPC/WORM client (de-dupe from merge2)
│  │  ├─ universal-sealer.js  # have
│  │  └─ receipt-drawer.js  # have
│  ├─ security/
│  │  └─ file-scanner.js  # have — 9-daemon swarm
│  ├─ media/
│  │  ├─ webcodecs-export.js  # have
│  │  └─ mp4-muxer.min.js  # have — local, no CDN
│  └─ ui/
│     └─ stack-panel.js  # have

├─ design-system/  # have — promote to first-class root
│  ├─ colors_and_type.css  # canonical tokens — never re-derive
│  ├─ ui_kits/  # app + marketing + _shared
│  ├─ brand-kit/  # logo · icon · swarm sigil
│  └─ preview/  # 29 design-system cards

├─ tests/  # CREATE — make the ephemeral proofs permanent
│  ├─ invisible-mark.spec.mjs  # BER / PSNR / wrong-key
│  ├─ file-scanner.spec.mjs  # malicious-file corpus
│  ├─ kem-envelope.spec.mjs  # round-trip + tamper
│  └─ fixtures/  # clips · attack samples

├─ dist/  # built portable artifacts + WATERSEAL.json
├─ infra/  # epoch-core-worker + flash-sync (local agent's domain)
└─ docs/  # consolidate
   ├─ HANDOFF.md  # have
   ├─ ARCHITECTURE.md  # CREATE — the layer map, in prose
   └─ THREAT_MODEL.md  # CREATE — what the marks/seals do and don't defend
04 · GAP ANALYSIS

What's missing, in priority order.

Ranked by leverage. The verifier and the launcher are the two highest-value additions — a provenance product that can mark but can't verify is half a product, and a suite with no front door reads as scattered scripts.

P1 · core

Verifier surface

The whole platform marks & seals but has no first-class tool to verify: drop a file + its .epochseal → SHA + signature verdict; drop a watermarked video → recover the invisible mark. Verification is half the value of provenance — and the marketing kit already promises a "public verifier."

apps/verifier/
P1 · core

Unified launcher / front door

Watermarker, sealer, merge ×5, verifier live as scattered HTML files. An enterprise suite needs one home that routes to every tool, carries the brand, and shows the current build/seal. Today there is no single entry point.

apps/launcher/
P1 · crypto

ML-KEM crypto bundle

kem-envelope-client.js matches your canonical FIPS-203 format exactly but is inert until the vetted epochcore-pqc-crypto.js (noble, offline esbuild) lands. One local build command unblocks recipient-encrypted .epochcrypt.

packages/crypto/
P2 · debt

Tools are off the design system

The watermarker and merge tools use an ad-hoc gold #c9a86a / #0c0d10 palette — not the canonical Swarm Violet #7d5bf4 App face. They look like different products. Reskin to colors_and_type.css tokens.

apps/*
P2 · debt

Duplicated engine modules

epochcore-api.js, receipt-drawer.js, webcodecs-export.js are copied into both epochcore-watermarker/ and merge2/. Promote to packages/ and import once — a single source of truth per module.

packages/
P2 · trust

Tests are ephemeral

Every BER/PSNR proof, scanner attack run, and envelope round-trip lived in throwaway scripts. For an enterprise crypto product the proofs must be permanent, re-runnable artifacts in tests/ — they are the evidence layer.

tests/
P2 · feature

Audio-only watermark

The invisible mark is luma-domain (video/image). Audio files currently ride the video path. The "mark any media" promise needs a real audio spread-spectrum mark for standalone .wav / .mp3 / .flac.

packages/crypto/audio-mark.js
P3 · robustness

Tier B re-encode survival

The invisible mark survives the tool's own encode (Tier A, proven). Surviving a hostile third-party transcode needs the WASM-DCT + sync-template upgrade. Right call for the entertainment/social subset, additive on the proven base.

packages/crypto/
P3 · polish

Geist wordmark font

The BrandMark wordmark calls for Geist; it currently renders in system-ui fallback. Drop in the TTF/WOFF to close the one known brand-fidelity caveat.

design-system/
atlas  EpochCore Platform Atlas · Quantum Watermarking
scope  4 tools · 11 engine modules · 2 UI kits · 1 dataplane
build  wm-3ae92a09d0cf · 2.0.0-universal-seal
sha-256  3ae92a09d0cfde69ad79394c9e2436b61dcd5a5b451a36890a9c867c6650bedd
ras  40668c787c463ca5  ·  coherence 0.9978  ·  freq 1210.0 Hz
✓ FISHBOWL MAPPED
© 2026 EPOCHCORE LLC · ALL RIGHTS RESERVED · PATENT PENDING · SEALED · sh.E.waterseal/v1 · strategic atlas · not a deploy