The Wargame Extension
An optional Strategy-Day layer — role-scoped cells, phased turns, an AI Adjudicator drafting consequences with faculty as umpire, and a signed After-Action Report.
An optional layer on top of the core platform (doctrine Q4 — wargame is a tenant extension, not core). Enabled per project via the Sessions + Workshop surfaces. Public page:
products-wargame.html. Marketed as a turnkey hybrid-warfare Strategy Day (HSG canonical case, CHF 40k fixed / 4-month build).
Concept
A "simulation" in insigz isn't a physics engine — it's structured human play against the real live world model, with an AI Adjudicator drafting consequences and faculty as the umpire. insigz already models the world the exercise simulates (live cables, vessels, grids, sanctions, news); the wargame adds a multiplayer layer with roles, phased turns, structured actions, real-world injects, and an automatic After-Action Report. Following RAND 2024, every AI output is a transparent suggestion with visible reasoning; faculty stays the authority.
The four deliverables of a Strategy Day
- Canonical scenario — one deeply-rendered scenario (the canonical example: Baltic submarine-cable cut at winter peak). Custom scenarios at additional scope.
- Role-scoped cells — each team sees only its own world; information asymmetry enforced at the database, not by trust.
- Adjudicator + injects — AI suggests consequences with full reasoning; faculty approves/edits/overrides every output.
- Signed AAR — delivered in five working days; becomes the next class's reading.
Cells (role-scoped teams)
- NATO / EU Cell — coordinates political-military response (national intelligence, operator reports, sanctions levers).
- Hybrid Threat Cell — pursues confidential objectives (threat plan, shadow-fleet positions, deniability budget).
- Civil / Media Cell (optional third) — public information environment (press lines, narrative pressure, civilian impact).
Preview each cell's world on Team view (/team-view).
Phased turns (the canonical scenario clock)
| Phase | Time | What happens |
|---|---|---|
| Phase 0 — Briefing | 45 min | Faculty briefs all teams on the world state; cells read dossiers and prepare opening actions. |
| Phase 1 — Initial response | 60 min | Hours 0–12. Each cell submits 2–3 actions; the Adjudicator drafts consequences (faculty approves); two scheduled injects fire. |
| Phase 2 — Escalation & attribution | 90 min | Hours 12–72. 4–5 actions per cell; attribution becomes central; three injects; optional team-to-team back-channels. |
| Phase 3 — Resolution & debrief | 60 min | Hours 72–168. Final positions; faculty narrates the verdict; 20-min structured debrief. |
≈4.5h play + 45m briefing + breaks ≈ 6h = one Strategy Day.
The Workshop (/workshop) — scenario authoring
Three editors (each an editable table/list with EDIT modal / DEL / + NEW, toast feedback):
- Action categories — what each cell may DO (e.g. Cable Inspection, Sanctions Designation, Article 4 Consultation, Information Operation, Cyber Pre-staging, Shadow Fleet Reposition, Public Press Line), each with cell, cooldown, cost, risk.
- Adjudicator playbook — how the AI drafts consequences (triggers → reasoning patterns).
- Inject library — real-world events fired during play, each tagged with the alert type it flows into (doctrine Q6 — injects surface in the 🔔 inbox in-product, no standalone inject surface).
Sessions (/sessions, /sessions/:id)
The session list (id · name · status · date · cells · players · AAR) and the session detail (status banner · phase timeline · cells · live faculty queue). Statuses: SCHEDULED / LIVE / COMPLETED. A LIVE session shows pending Adjudicator consequences in a faculty queue; a COMPLETED one links to its AAR. Seed sessions include BALTIC-2027-0217 (completed, AAR published), BALTIC-DRESS-0214 (dress rehearsal), and HSG-WG-0319 (live, Move 2, AAR drafting).
AAR drafting (/aar-drafting)
The staff after-action surface. Sections: Timeline · Decisions · Base rates · Prompts (NARRATIVE) + a Reasoning appendix (APPENDIX, §-divided chains). Each section approve/un-approve; counter N/5 APPROVED. Q8 export gate: PDF export locked until all five are human-approved → then SHIP TO CUSTOMER →. PREVIEW PDF opens a print-styled A4 window (running headers/footers, page-break + orphan/widow control, signed-by footer) and triggers print.
Faculty time commitment (the engagement shape)
- 4 weeks before: scenario authoring (~10h over 4 sessions) — objectives, role briefs, curriculum themes.
- 1 week before: dress rehearsal (one 2h session; faculty plays umpire, insigz plays the cells).
- Strategy Day: ~6h on-site facilitation (faculty teaches, insigz operates).
- 1 week after: debrief + AAR review (~1h) — faculty edits and signs.
Engagement (from products-wargame.html)
Commercials are presented as "on request" (fixed price / fixed scope / fixed timeline; no figures published). A first engagement is one Strategy Day on one canonical scenario; repeat engagements in later semesters reuse the platform investment. Optional add-ons (scoped per engagement): custom scenario authoring beyond the canonical one, a third cell (Civil/Media) on top of the default two, multilingual AAR delivery. Term is a fixed 12 months, multi-year optional. No named institution or fixed delivery timeline is quoted on the public page.
How to run a custom scenario (e.g. "surveillance progress since 2023")
A wargame points forward; "since 2023" points backward. Do both: build a Case + AsOf reconstruction of the historical trajectory as the briefing dossier, then start the session clock at "today" and let the cells play forward.
The Ontology Agent (source onboarding)
The chat-driven 8-stage wizard that profiles, connects, maps, and dry-runs a new source — the agent drafts the canonical-model change, a human Steward commits it (Q9).
Settings & Admin
The two settings routers — organization settings (account, team, security posture, audit) and project settings (sources, members, modes & surfaces, danger zone).