Oshun V1 — Features (Page Index)

This directory holds the Oshun V1 feature map, decomposed from the single ../features.md hub into focused, in-depth pages. The hub remains the canonical entry point and preserves every heading anchor (so existing features§"…" cross-references keep resolving); each page below expands one section with far richer, code-grounded prose.

Release scope (2026-08-05). This index covers the whole V1 line, not a V1.0 shipping list. V1.0 ships the customer web app + PWA with four rooms — Tara, Nyx, Arete, Nisaba — charging through the crypto rail only. V1.1 adds the native mobile apps and the fiat payment rails. V1.2 opens the Veritas and Metis rooms. Pages for deferred work carry a release banner at the top; the scope record is ../BRAND.md.

Companion documents: the runtime/package architecture lives in ../architecture/, the backlog (source of completion truth) in ../TODOS.md, the dependency registry in ../DEPENDENCIES.md, and the 2026-06-24 review/audit ledger in ../AUDIT_2026-06-24.md.

Cross-reference convention. A bare §N / §N.M points at a section of the backlog ../TODOS.md; deps§N points at ../DEPENDENCIES.md; page-to-page links are relative (./<slug>.md, or ../architecture/<slug>.md for the companion).

Feature landscape#

This is a navigation model, not a dependency graph. The architecture index owns runtime composition; the release table and per-page banners decide which parts of this feature landscape are member-visible in each V1 release.

Orientation#

• V1 Product Promise — the experience V1 commits to across customer, admin, assistant, and platform surfaces.
• Subsystem Glossary — domains, substrates, surfaces, and cross-domain support defined in one place.
• Product Surfaces — Customer Web/Mobile, the shared consumer shell, operator surfaces, and public web/distribution.
• Account, Profile, and Settings — the account hub: identity, preferences, privacy/consent, data rights, devices, memory controls, and safety settings.

Customer Domains#

• Tara — Rituals and Contemplative Practice — rituals, breathwork, meditation, teachers, and contemplative continuity.
• Arete — Goals, Habits, and Reflection — goals, habits, journaling, weekly review, and humane streak/recovery coaching.
• Veritas — Grounded Stories and Claims (V1.2) — claims, sources, evidence, counterclaims, topic hubs, timelines, and the retraction cascade.
• Nyx — Sky Events and Perspective — sky events, observation windows, awe/perspective, and calendar integration.
• Nisaba — Scholarly Study — passages, manuscripts, editions, translations, lexicon/morphology, the concept graph, notebooks, study plans.
• Metis — Education and Tutoring (V1.2) — courses, tutoring, bring-your-own-material study, assessments, and standards-based institutional delivery.

Authoring, Editorial, and Curation#

• Creator Roles and the Authoring Workspace — contributor roles and the core authoring surface.
• The Tara Content Workbench — the Tara meditation-content pipeline: spark → gated draft → approval ladder → validated publishing bridge into apps/tara/content.
• Editorial Calendar and Asset & Media Library — the editorial lifecycle and the asset/media library.
• Taxonomy, Localization, and Versioning — concept-graph curation, the translation workspace, and diff/rollback.
• Collaboration, Review, and Templates — comments/review threads and the templates library.
• Customer Curation, Notebooks, Collections, and Sharing — customer-side notebooks, collections, and sharing.

Assistant, Memory, Persona, and Embodiment#

• Assistant Experience — the unified assistant surface and its entry points.
• Iris Memory and Identity — memory schema, recall mechanics, consent, suppression, deletion/export, and admin inspection.
• Psyche Real-Time Runtime — turn-taking, interruption, transcript sync, lip-sync, and reconnect continuity.
• Lilith Persona Policy — tone/teacher-persona safety, crisis policy, and voice-abuse handling.
• Persona, Avatar, and Voice Packs — the persona/avatar/voice registry, approval workflow, and policy gates.
• Mentor Presence — Meditating with a Master — the embodied-mentor layer for the meditation player: governed mentor avatars that arrive, settle, and guide the sit (§36; implemented, release-gated).

Grounding, Generation, Media, and Search#

• Sophia Grounding — grounded answers, evidence packs, citation trails, and contradiction loops.
• Isis Generation Control — the governed generation control plane.
• Generation Audience Tiers and Surface Boundaries — tiered generation entitlements and surface boundaries.
• External Model Intelligence and Execution Providers — Civitai intelligence and ComfyUI/RunPod execution.
• Creator Surfaces, Voice, Music, and 3D Generation — curated creator surfaces and voice/music/3D generation.
• The Atelier — Five Rooms of Making — the customer generative workshop: five rooms, three house rules, the casting seam, and member scenes.
• Output Gallery, Lineage, Branch, and Replay — the generation gallery, lineage graph, branching, and replay.
• Search, Discovery, Recommendations, and Knowledge Graph — search, discovery, recommendations, and the knowledge graph.

Living Scenes#

• Living Scenes — Concept and Customer Promise — what a Living Scene is and the promise it makes.
• Scene Score Schema — the declarative scene-score contract.
• Live Direction, Conductor Runtime, and Blend Kernel — the real-time direction channel, conductor, and blend kernel.
• Composition Surface and Cinematographic Technique Catalog — the composition surface and technique catalog.
• Domain Templates and the Scene Score Editor — per-domain templates and the AAA scene-score editor.
• Scene Safety, Determinism, Provenance, and Cue Privacy — safety/tone/crisis, re-render determinism, C2PA provenance, and cue redaction.
• Keep, Share, Shareability, Takedown, and Lineage — keep/share, the shareability matrix, takedown cascade, and remix lineage.
• Latency, Accessibility, Eval Sets, and Tests — latency/backpressure budgets, reduced-motion renders, eval sets, and tests.

Agentic AI Studio#

• Agent Registry, Job Orchestration, and Multi-Agent Plans — the agent catalog, long-running runs, plans, and operator dashboard.
• Agent Invocation, Budgets, Memory, and Feedback Loops — customer/creator invocation, budgets/kill-switches, memory, and gold sets.
• Tool Catalog, Grant Semantics, and Multi-Agent Protocol — the tool catalog, grant semantics, and the hand-off protocol.

Admin, Tenant, and Operator#

• Admin Products — Web and Mobile — the operator admin web and mobile products.
• Tenant, Institution, and Operator Toolkit — tenant/org admin, SSO/SCIM, roles, audit log, bulk ops, API keys, comms, help center, status.

Governance, Safety, Support, and Privacy#

• Review, Compliance, and Trust & Safety — review/audit/compliance/rights/provenance and the trust & safety operations.
• Support, Entitlements, Billing, and Customer Operations — support tooling, entitlements, and billing operations.
• Crypto Payments — Non-Custodial Entitlement Settlement — the non-custodial crypto settlement flow and entitlement bridge.
• Privacy, Consent, Data Portability, and User Controls — consent, portability, DSAR, and user-facing privacy controls.

Messaging Channels#

• Channel Abstraction, Routing, and Boundaries — tier-aware channel routing and the residency/retention/consent boundary.
• Telegram Surfaces — Bot, Mini Apps, Channels, Inline, Auth, Payments — the full Telegram surface set.
• WhatsApp, Email, Push, SMS, Discord, and Slack — the remaining transactional and community channels.

Platform, Quality, and Launch#

• Architecture, Platform Foundations, and Security — the platform foundations and security/operational hardening.
• Design System and Experience Quality — the design system and experience-quality bar.
• Analytics, Observability, Testing, and Security — analytics/experimentation/evaluation and the testing/QA/a11y/perf/security posture.
• Content, Localization, Documentation, Launch, and Exit Criteria — content/localization/docs, launch, and the V1 exit criteria.