FocusForge

Distributed-narrative cast

Meet the cast

FocusForge's 6-character cast embodies EXECUTIVE-FUNCTION DOMAINS as CAPACITIES-not-GAPS: Hold (working memory) / Wait (inhibitory control) / Pivot (cognitive flexibility) / Map (planning + organization) / Begin (task initiation) / Clock (time awareness). Anchor (the existing AI mentor) introduces and references each EF domain. ADHD-shame gate enforced (research-cited; see /method): cast NEVER says 'you should focus better' / 'try harder' / 'stop being distracted'; cast frames every skill as a CAPACITY to BUILD never a GAP to FILL; NO before/after character pairs (broken-clock / scrambled-brain visuals REJECTED); neurodivergent-affirming voice baseline ('ADHD is a way your brain works, not a flaw' / 'all brains have to learn these skills'); static-response gating for shame signals ('I can't focus' / 'I'm broken' / 'my brain doesn't work') NEVER reaches FoundationModels — Anchor surface filter routes to neurodivergent-affirming reply; kit 6 + kit 10 carry off-ramp affordances per SAMHSA TIP 57; no second-person imperative catchphrases ever — all first-person modeling. KitMetadata.castAllowed flag required for any DIR-FEDC-capacity-1-sensitive kit; cast pacing respects DIR-FEDC attention duration. Soft collision with ReadQuest Anchor cast member allowed (mentor vs cast; different domain). A second, deeper layer of four friends goes further inside the work itself: checking as you go, staying through the long middle, choosing what comes first, and chipping through the hard part one piece at a time.