← Return to Wolfberg Wolfberg WOLFBERG PER ASPERA WAR ROOM · AWS Global Government Hackathon · Jun 29 - Jul 1, 2026
ONE GOVERNED RUNTIME · FIVE MISSIONS · ONE OPERATING MODEL

You don't need more AI tools. You need to solve problems at the speed of the fight.

Two days. 2 prototypes, 3 MVPs, 8 PoCs, 8 conceptualized wireframes. The AWS infrastructure they all can share. The tamper-evident ledger tracking every human-machine interaction.

We didn't buy more AI tools, we learned how to use the ones we had.

The operating model IS the accelerator.

The frame · how to read every mission on this site

Your AI isn't a tool. It's your newest employee.

The biggest question in AI isn't which tools are best. It's the one everyone keeps asking: how do we use this? The answer is not more AI tools. It is leading the ones you already have.

So what

Huge speed and efficiency gains in digital problem solving. The operating model took a full production platform from nothing to launched in 266 logged hours, counted, not estimated. The same scope, built the conventional way, models to 8,000 to 21,000 engineer-hours (Brooks team-sizing). Here you watched the same model stand up 21 mission builds, five live and governed on one engine, in two days.

01 · Onboard

One shared brain, as code

Every agent works from the same context, memory, and rules, kept as code. Train it once; it stays trained.

02 · Scope

Rules that keep it honest

Every claim traces to an artifact. Nothing unsourced ships. A health check pulls any agent that drifts.

03 · Supervise

A human leads, all logged

A human approves every irreversible call. The decision signs into one hash-chained ledger you can replay.

Three doors

Judge walkthrough first, the live consoles second, the verification surface third.
JUDGE WALKTHROUGH · START HERE
10 minutes, in order

The sequence-of-eyes for judges: README, the 90-second receipts, the live consoles, the credibility receipts. Lands the load-bearing claims without missing the receipts.

Open the judge walkthrough →
C2 CONSOLE · FIVE MISSIONS, ONE ENGINE
Multi-mission operator picture

Switch between #01 / #02 / #04 / #07 / #08 on the mission rail. Same gate, same signed ledger, same always-live abort underneath. Per-mission SCOPE OF CLAIM panel travels with the demo.

Open the C2 console →
C-UAS 3D · THE DEEP PROOF
The same governed engine, in 3D

The Counter-UAS floor, rendered as a live 3D battlespace. Real RF / radar / EO-IR detection geometry, drones on real tracks, an LLM-free threat call on each, and a recommend-only DEFEAT held behind a two-person gate. Same gate, same signed ledger as the C2 console — no separate build, no fork.

Open the C-UAS 3D →

The thesis

How the speed and cost above become a runtime a government can actually field.
We are not shipping five demos. We are shipping one governed runtime with a human gate on every irreversible call and one signed ledger, and behind it the operating model that built it. Five missions prove the runtime generalizes. The two days above prove the model does.

The deep proof is Counter-UAS #08, the 1 km² defended area, real C-UAS track fusion through a live stream, the graduated-authority gate with positive hostile-ID at weapons-tight and a two-person rule, effector tasking under BSI Flex 335 / MITRE CoT (simulated). #07 FOB UAV/UGV is the same engine, the discrimination win: a real ADS-B Cessna 172 (ICAO a24c5a) correctly classified cooperative against the synthetic drone threat. #01 seabed, #02 convoy, and #04 swarm-recon are honest width on the shared spine: real corpora, governed gate, signed ledger, mission posture stated on each card.

Open: Judge walkthrough Deep: #08 + #07 on the C2 console Throughout: Signed ledger accumulating
⚠ The trap

The rubric front-loads Working Prototype. One running thing beats four mocks. We win on the depth of the governed loop, not the breadth of the graph. The line we will not cross: selling maritime, seabed, and swarm at parity to C-UAS is overclaim. We do not.

The five missions

One engine on top, five domains underneath. Only the fusion stage and the corpus differ. Per-mission scope-of-claim travels with the console card.
Challenge #8Deep proof

Area Counter-UAS

The 1 km² defended area. Radar + RF + EO/IR fused tracks, IMM trajectory prediction, NINJA-EW + LEONIDAS-HPM under BSI Flex 335 / MITRE CoT (simulated effector). Two-person auth required at engage; RF NO-EMIT sector; graduated ROE.

12 / 12 answer-key invariants reproduced · live on real drone-track data
Challenge #7Discrimination win

FOB UAV/UGV Defense

Same engine as #08, no fork. Real ADS-B cooperative anchor (Cessna 172, ICAO a24c5a, adsb.lol open API) correctly classified cooperative against the synthetic drone threat. Real cooperative aircraft, distinguished from a threat, on the same gate.

13 / 13 answer-key invariants reproduced · import-graph shows no fork
Challenge #1Honest width

Seabed Cable Corridor

Surface-AIS picture over a public submarine-cable landing area (San Pedro Channel). The shared gate and ledger under UNCLOS recommend-only posture. Subsurface acoustics deliberately excluded; the metamorphic harness proves no answer-key leak.

real NOAA AIS · 60 vessels / 6,775 fixes · signed ledger verifies
Challenge #2Honest width

Convoy Escort

Real Marine Cadastre AIS (San Pedro Bay), same governed gate, signed ledger. Cooperative AIS gets HOLD_FRIENDLY; hostile or suspect gets RECOMMEND under UNCLOS posture. Same engine as #01, different mission posture.

real NOAA AIS · 65 signed-ledger records · envelope as entry #1
Challenge #4Honest width

Swarm Reconnaissance

Mapping mosaic, not threat tracking. Real SkyFusion overhead-imagery detections (Apache-2.0, annotations only) driving a mosaic-fusion variant on the same gate and ledger spine. COP-commit is human-gated; the gate ladder differs from #08 by design.

real SkyFusion imagery · metamorphic harness green · one-engine proof

The architecture

One governed runtime, edge to cloud. The disconnected leg runs forward; the cloud leg is reached on reconnect.
● Edge · disconnected
  • AWS Snowblade (DoD / JWCC) for the forward node, or
  • IoT Greengrass v2 + SageMaker Neo small models on the device
  • Sensor fusion, detection, and the gate run with no reachback; deltas buffer for reconnect
● Cloud · GovCloud, on reconnect
  • Bedrock + AgentCore are cloud-only; reached when the link is back
  • S3 Object-Lock WORM anchor for the ledger; reconcile on reconnect
  • Cross-site intelligence and after-action records consolidate here
Contract: /contracts pipeline-events v0.1.0. One unified envelope. CoT-compatible payload for tactical missions, ASFF + NIST 800-53 r5 mapping for security, non-biometric shapes for sensors.
Governance, cross-cutting: the deterministic gate, a human approves every irreversible call, one Ed25519-signed hash-chained audit ledger across every mission. The LLM is off the kill chain — the no-model-call CI test on the gate module is the proof, plus the IAM-denied narration role at the infra layer.
Do not cite: GPU Snowball Edge and Snowcone are discontinued. The disconnected story is Snowblade or Greengrass v2 + Neo. Deployment is commercial AWS (us-east-1), not GovCloud; GovCloud IL5 is a clean target mapping, not the current boundary. Classified and PII are out of scope.

Team and lanes

Your lane and your first task. Instances and people coordinate through Berg or the bus, never directly.
Berg
Conductor + gate
Lane
Own the demo spine: judge walkthrough → C2 console (2D + 3D) → signed ledger accumulating throughout. Gate every PR, publish, and deploy.
gates all PRs / publishes / deploys
Charles Gibson
StackWright / Per Aspera
Lane
The multi-sensor pipeline and the governed-run monitor. ESF-8 Storm Surge as a second-domain proof. Apps/monitor renders the signed event stream off the pipeline-events envelope.
apps/monitor · ESF-8 Storm Surge subsystem
Maryam Shahbaz Ali
Quality & trust — cross-cutting
Lane
Evaluation & verification — the metamorphic invariance / LLM-judge harness, the gate-decision explainability and audit, and the after-action narration eval (R-GC3).
metamorphic harness · explainability + audit
Clyde Smithson
Systems architecture & integration
Lane
System-of-systems coherence across the five missions. The one-engine import-graph proof, MBSE and VV&A rigor, and C2 / battle-management depth across the mission family.
one-engine import-graph proof · MBSE / VV&A · C2 depth

How to start

Clone, bootstrap, verify the contract loop, then drive the missions.
wolfberg-peraspera-hackathon - bash
$ git clone <wolfberg-peraspera-hackathon>
$ cd wolfberg-peraspera-hackathon
$ ./onboarding/bootstrap.sh                          # corepack + pnpm install + typecheck + contract loop
# expect: OK — contract loop verified.
$ pnpm --filter @wolfberg/c2-mission demo             # #08 floor: real drone-track fusion + gate + signed ledger
$ pnpm --filter @wolfberg/c2-mission demo:07          # #07 FOB reuse: same engine, swap config
$ pnpm --filter @wolfberg/c2-width demo               # #01 + #02 maritime: real NOAA AIS + governed gate + ledger
$ pnpm --filter @wolfberg/c2-fusion test              # #08 12/12 + #07 13/13 answer-key invariants
/contracts pipeline-events /core governed runtime /packages/sdk /packages/mock /solutions/c2-fusion /solutions/c2-gate /solutions/c2-mission /solutions/c2-missions /solutions/c2-width /apps/c2-console /datasets /infra /.kiro /onboarding
Status check: if the bootstrap loop does not print OK — contract loop verified., stop and ping Berg. Do not improvise around a missing contract loop. Build state can lag this page; main is the source of truth.

Schedule and constraints

Hard rules and the build receipts. Ship-Reality holds on every surface.

Constraints (hard)

  • AWS Kiro mandatory for the build
  • Real corpus + disclosed synthetic scenario per mission. No classified or PII inputs (MMSI hashed at the adapter; SkyFusion annotations only)
  • Commercial AWS, us-east-1. Not GovCloud; classified and PII are out of scope
  • Event: AWS Global Government Hackathon. Judging Wed 2026-07-01

Ship-Reality (non-negotiable)

  • Running prototype on real open-source data with the governed loop. Not a deployment, not an ATO.
  • No prompt-caching cost claim. Not wired; do not put a caching number on any surface
  • NIST 800-53 = Rev 3-class self-assessment, r5 mission mapping. Never say "compliant"
  • Effectors are simulated. Nothing actuates. The ledger records what would have been dispatched
Cost posture and the build receipts.
~$1/day
Edge run cost · with a one-time ~$3 model pass
5
Missions on the shared spine
TRL 4 (partial 5)
Validated beyond a lab on real open-source data with the governed loop. Not operationally demonstrated
Deploy posture: the war room runs on Wolfberg's own AWS (CloudFront, on a wolfberg.ai subdomain). Reproducible deploy via scripts/deploy-warroom.py --apply. Berg-gated apply.