RWF — Timeline & Achievable Goals

2 Sep 2026 · v1.1 baseline. The plan pairs with the contract milestones (docs/RWF_Contract_Scope.pdf) — dates assume Ben engages this week.

Where we are (v1.1, today)

Done: engine + parity tests · v1 app (65 screens, power-ups, DZ, daily winner, seasons) · v2 board-game app (poker-table × track, cards, kitty, 13 themes) · points currency + draft-from-3 + reroll-to-kitty + catch-up curve · multi-squad dashboard + wagers · SFX layer · WhatsApp/Slack bots · Beeper transport (needs token) · Android app on the phone · avatar systems (Geno mocap, wardrobe, frogs, anyCreature, img2threejs foundry verdict) · public coverage hub (rwf.qalarc.com/v1).

Weeks 1–2 — M1: One system (contract Milestone 1)

[ ] App↔API↔bot state unification (T5) — apps/api is the single store
[ ] Magic-link auth (Resend) — identity across phone/bots
[ ] v2 board app as the primary client (v1 archived, kept for progression)
[ ] Re-sync the v2 engine fork with the draft/points API (banner'd TODO)
Goal: a crew created on a phone plays a match visible from a WhatsApp group — no orchestrator.

Weeks 3–4 — M2: Always-on + real verification (Milestone 2)

[ ] Bots on GMKtec 24/7 (Slack tokens → ~/.config/rwf/; WhatsApp via Hub tunnel or Cloud API decision)
[ ] Beeper live (founder mints token — 3 min) → Telegram/Discord reach free
[ ] Camera verification hardened with 3 real users (MoveNet accuracy log)
[ ] Cross-squad log dedupe rule (server-side fairness)
[ ] Sound pass on v2 (the SFX module is wired; tune the catalogue)
Goal: 72h uninterrupted, 2 groups playing without intervention.

Weeks 5–8 — Pilot (contract Pilot milestone)

[ ] First 3 real crews (mates + one corporate-friendly group)
[ ] The 5 product decisions implemented per Ben's calls (fairness, platform, rhythm, power-ups-in-MVP, brand — docs/24 §9 has both directions ready)
[ ] Weekly digest shipping to real chats
[ ] Points economy tuning: reroll costs, catch-up curve steepness, kitty flow — from real play data
[ ] Avatar identity pass: squad tokens → generated avatars (img2threejs foundry lane) or species-as-tier (frogs)
Goal: 3 groups, ≥2 matches/week each, W2 retention ≥60%.

Weeks 9–12 — Post-pilot: the money + the store

[ ] Points→money decision from pilot data (buy points after the trial verdict — the founder's stated sequence)
[ ] Charity pot legal opinion → employer-funded pots first
[ ] WhatsApp Cloud API migration (groups verdict) or per-player DMs
[ ] Corporate pilot (Slack org install — the revenue engine)
[ ] Public launch ready: store listing, the Silverchair publicity wave landing on /v1 + a public ladder (element C17)

Standing goals (every phase)

All suites green on every push (currently: 49 engine + 122 + 74 + 90 + 56 board + 98 app + bots + styles — ~600 checks)
One new real-user insight captured per week (the wiki's ops page is the log)
The blockers register (docs/15) reviewed weekly; nothing red for >2 weeks

What would make each phase FAIL (be honest early)

M1: state unification sprawls → cut v1 parity, ship v2-only
M2: Hub uptime → WhatsApp Cloud API decision early
Pilot: nobody plays → the handicap/comeback numbers need tuning from figma-app telemetry; power-up economy rebalance
Post-pilot: legal stalls charity pots → employer-funded pots are the bridge

UPDATE 2 Sep (evening) — new points recorded from founder directives

v2 IS the direction: track-and-field board game, REPS language only ("the pot", never kitty; poker words swept — style references only)
Themes overhauled to true design languages (8 rebuilt + app-screen demos at /styles §03 "On the app")
Main page = the FULL HUB (version links, SFX demo at /sfx, everything)
v1 preserved forever at /figma-app (design progression)
New standing directives live in agents/ORCHESTRATION.md — that file is the contract for every future agent