Skip to content

Gameplay Wireframe Audit — 2026-04-21

CIPHER-LINE revision note (2026-04-24): Where this audit discusses the “Cipher feedback line” or “Cipher narration,” read those as the CIPHER-LINE OLED output surface (not the main 80×25 grid). Marty Glitch’s “Cipher line replacement” is explicitly a CIPHER-LINE takeover via anti-grammar contributions — see docs/gameplay-specs/KN-86-Marty-Glitch-Broadcast-Piracy-Gameplay-Spec.md Parasitic Override section and docs/architecture/KN-86-CIPHER-LINE-Grammar-Framework.md.


Good news: The gap is much smaller than anticipated. The kn86-attract/src/scenes/library/ directory already contains a near-complete set of animated Remotion scenes — 14 of 16 spec modules are fully designed with the canonical 7 scenes + 2 sequence configs each.

The only outstanding gap is Marty Glitch. The NeonGrid-BlackLedger combined spec is a re-documentation of two modules that already have their own standalone scene sets (so it does not require its own fresh wireframes).

Static PNG wireframes in kn86-screens/ remain partial (only 4 modules: ICE Breaker, Depthcharge, NeonGrid, BlackLedger). That older gallery is now a subset of what the Remotion scene library covers, and rendering frame-captures from scenes into that folder is an optional follow-up rather than a blocker.


Two wireframe surfaces were inspected:

  1. kn86-screens/ — static PNG mockups, NN-module-screen-name.png convention.
    • 01..10 = OOBE / boot / bare deck / SYS (firmware, not module-owned)
    • 11..17 = ICE Breaker (7 PNGs)
    • 21..27 = Depthcharge (7 PNGs)
    • 31..37 = NeonGrid (7 PNGs)
    • 41..47 = BlackLedger (7 PNGs)
  2. kn86-attract/src/scenes/library/ — TypeScript Remotion scenes (the authoritative wireframe artifact per the kn86-scene-designer skill).
    • Each fully-designed module has 7 scene files following the {module}-{screen-name}.ts convention plus 2 sequence configs in kn86-attract/src/sequences/.

Spec screen counts are not a hard gate — the scene-designer skill standardizes on a fixed 7-scene arc (cartridge load → primary gameplay → secondary mechanic → tool use → tension → success → failure). Each gameplay spec lists more than 7 candidate screens; the 7-scene arc selects the visually-distinctive moments.


#ModuleStatic PNGs (kn86-screens/)Scenes (kn86-attract/src/scenes/library/)SequencesGap
1BlackLedgerYes (7: 41-47)Yes (7)Yes (2)None
2CipherGardenNoYes (7)Yes (2)PNG export optional
3DepthchargeYes (7: 21-27)Yes (7)Yes (2)None
4DriftNoYes (7)Yes (2)PNG export optional
5ICE BreakerYes (7: 11-17)Yes (7)Yes (2)None
6Marty Glitch — Broadcast PiracyNoNo (0)NoNeeds design
7NeonGridYes (7: 31-37)Yes (7)Yes (2)None
8NeonGrid-BlackLedger (combined spec)N/ACovered by standalone modules (7+7)N/ACombined spec is re-documentation; no new scenes needed
9NodespaceNoYes (7)Yes (2)PNG export optional
10NullNoYes (7)Yes (2)PNG export optional
11PathfinderNoYes (7)Yes (2)PNG export optional
12RelayNoYes (7)Yes (2)PNG export optional
13ShellfireNoYes (7)Yes (2)PNG export optional
14SynthFenceNoYes (7)Yes (2)PNG export optional
15TakezoNoYes (7)Yes (2)PNG export optional
16TheVaultNoYes (7)Yes (2)PNG export optional

Legend: “Gap” = what is genuinely missing. “PNG export optional” = scene files exist; a frame-capture export into kn86-screens/ would complete the static gallery but is not required for design integrity (the scene files are the authoritative artifact).


Marty Glitch — Broadcast Piracy is the only module without any wireframes in either location.

Marty is an unusual capability module: it is a parasitic overlay cartridge that hijacks the Cipher feedback line and occupies the right half of the screen (cols 40-79, rows 1-20) with an animated glitch face while the base cartridge’s mission UI runs on the left half. It also contributes three of its own mission templates:

  • MT_FIND_GHOST_SIGNAL (spectrum scan / frequency triangulation / signal lock — CAR/CDR/CONS band navigation)
  • MT_LIBERATE_JAMMED_FEED (multi-layer decryption / jamming bypass / extraction — CONS/LAMBDA/APPLY stacking)
  • MT_HIJACK_CORPORATE_BROADCAST (payload injection / hold duration / real-time ICE evasion — CAR/CDR/QUOTE channel management)

Plus Marty introduces two cartridge-local systems:

  • Signal Integrity (0-100, degrades with use, corrupts the animation layer)
  • TT Standing (Off-the-Grid reputation, parallel to Universal Deck State)

And a unique publisher splash (SETEC ASTRONOMY → TOO MANY SECRETS anagram, 3s / 90-frame).

The 7-scene arc that fits this module cleanly:

  1. Cartridge Load — SETEC ASTRONOMY splash (decoy / tear / reveal / restore), Cipher hijack, Marty face assembly, READY.
  2. Primary Gameplay — Mission board with Marty overlay (left: contract list; right: Marty face with ambient glitches; Cipher line replaced by stuttering pirate narration).
  3. Secondary Mechanic — Spectrum scanner / band navigation screen (MT_FIND_GHOST_SIGNAL representative — the most visually distinctive of Marty’s three templates).
  4. Tool/Capability Use — Archive browser for prior transmissions (CAR/CDR through archived broadcasts; Marty’s own EVAL to re-broadcast).
  5. Tension Moment — Corporate listener / trace escalation; Signal Integrity degradation visibly corrupts the animation layer.
  6. Success Extraction — Mission success; Marty’s success chime; TT Standing updates; archive entry recorded.
  7. Failure State — Failure buzz; Signal Integrity penalty; Marty’s commentary on the caught operator.

Estimated effort: Single module = 7 scenes + 2 sequences + library/Root registration, modest complexity. Within a single PR scope.

Soft gap (static PNGs are partial, not a blocker)

Section titled “Soft gap (static PNGs are partial, not a blocker)”

10 modules have Remotion scenes but no static PNG wireframes in kn86-screens/. This is a gallery completeness question rather than a design gap. Options:

  • Leave as-is. Scene files are the authoritative artifact.
  • Follow-up task: Render one representative frame per scene into kn86-screens/ using the Remotion pipeline’s still-frame capability. ~70 PNGs at a known prefix scheme (50s/60s/70s/etc.).

Recommend deferring to a separate task if Josh wants the gallery completed — this is export plumbing, not design.


Given the audit results, there is only one real design deliverable: Marty Glitch — 7 scenes + 2 sequences.

This fits cleanly in a single PR and stays well below the “halting rules” threshold in the task brief (which would trigger if >3 modules with >5 screens each remained). No scope-confirmation blocker is needed — the work is tractable and bounded.

Proposed Wave 1: Design Marty Glitch’s 7 scenes + 2 sequences per the scene-designer skill, register them in library/index.ts and Root.tsx, verify npx tsc --noEmit passes.

Static PNG export for the 10 modules with scenes-but-no-PNGs is recommended as a follow-up ticket if Josh wants the kn86-screens/ gallery to remain comprehensive. Not blocking GWP-147.