Privacy control
Wizzo uses essential cookies for login, security, and core app behavior. Optional analytics helps us find bugs and improve the product, and it only runs if you allow it.
Introduces Living Observatory as Wizzo's redesigned authenticated experience: a clearer Home with one authoritative Next Move, Focus for execution, Campaigns for long-term context, Mentor as the integrated AI workspace, Vault for memory/sources/privacy, and shorter first-run onboarding.
Read the latest releaseAdded an off-by-default, provider-independent decision-policy boundary across chat, Today/Journey, voice check-ins, and connector quest automation while preserving every existing production choice.
Read release notesAdded one persistent, privacy-safe recommendation-quality contract across chat, Today/Journey, and voice, including accepted-as-is, edited, rejected, dismissed, created, started, snoozed, completed, and handoff outcomes with normalized source attribution.
Read release notesSelected `v0.51.0` as the First-Win Follow-Through Beta, focused on turning one source-backed recommendation into one visible completed win across chat, Today/Journey, voice, personalization, and onboarding.
Read release notesHardened MCP connector availability so feature flags no longer present unconfigured connectors as usable, connector toggles reject missing runtime config server-side, and admin pages distinguish flag exposure from provider env/webhook registration.
Read release notesAdded the first v0.50 Today/Journey command-center slice: Journey now promotes one transparent Today plan with due, time, label, priority, source, and recent-behavior signals plus accept, snooze, complete, and chat handoff telemetry.
Read release notesAdded quest time tracking and privacy-safe labels as personalization primitives, including Trials create/edit fields, telemetry payloads, sequencing inputs, and personalization dataset exports.
Read release notesAdded opt-in MCP connector quest automation rules: connector events can now create review-first quest drafts or explicitly enabled auto-created quests with product-event and quest-audit provenance.
Read release notesAdded connector-aware smart quest suggestions so opted-in Calendar, Gmail, and Google Drive MCP activity can create attributed next-action prompts with acceptance/follow-through telemetry.
Read release notesReorganized backlog documentation so `docs/backlog/ACTIVE_BACKLOG.md` is the current execution queue, `docs/backlog/RELEASE_GATES.md` is the validation contract, completed release-prep and AI rebaseline artifacts live under `docs/backlog/archive/`, and sprint-history backlog redirect stubs are removed.
Read release notesClosed Milestone 10 sprint documentation by replacing Sprint 45 and Sprint 46 planning artifacts with canonical completion records, including a deferred closeout for mobile scope and a final connector closeout record for provider rollout and webhook audit delivery.
Read release notesStabilized first-message chat UX by preventing conversation bootstrap reloads from clobbering optimistic assistant streaming updates.
Read release notesHardened production API exposure controls with verified post-deploy endpoint gating and payload redaction for sensitive diagnostics and migration routes.
Read release notesPublished Sprint 44 closeout artifacts for Milestone 10 reprioritization continuity: `SPRINT_44_COMPLETION.md`, `sprint_44_master_prompt.md`, `SPRINT_44_TESTING_GUIDE.md`, and `SPRINT_44_TEST_REPORT.md`.
Read release notesSprint 43 (Milestone 9) closed with full phase coverage (0-8), including explainable autonomy decisions, follow-up suppression intelligence, and explicit user controls/kill-switch behavior.
Read release notes**Agentic tools phase 2 shipped:** code execution and image generation added to the explicit-invocation toolset with bounded outputs and visible result states.
Read release notes**Offline quest editing:** title, notes, and status changes persist locally and sync on reconnect with deterministic conflict handling.
Read release notes**Agentic tools phase 1:** web search, file uploads, and document analysis with explicit tool controls and structured outputs.
Read release notes**Cohort insights dashboard:** Journey AI Insights now surfaces cohort maps, cadence timing, and actionable behavior guidance based on clustering analysis.
Read release notes**Weighted quest scoring v2:** complexity, duration, difficulty, and theme importance multipliers with per-quest breakdowns and overrides.
Read release notes**Trial-scoped conversations:** create Trial-linked chats from nav and Trial pages with grouped navigation and filters.
Read release notes**Trial hero AI generation (flagged):** new AI hero studio flow with prompt presets, quota guardrails, and S3-backed storage (uses supported image sizes).
Read release notes**Trial analytics + comparison:** Trial progress dashboard, `/api/trials/[trialId]/progress`, and `/trials/compare` for side-by-side metrics.
Read release notes**Companion presence orchestration**: `PresenceProvider` and the animated `WizzoAvatar` now drive idle, listening, thinking, success, and warning states across chat, quest mutations, and navigation with reduced-motion fallbacks.
Read release notes**RL-driven inference caching**: Redis adapters now support an epsilon-greedy cache TTL policy with telemetry surfaced on the admin ML dashboard, improving hit rate and latency for personalization and pattern endpoints.
Read release notes**Multi-channel notifications shipped**: Push/email/webhook preferences, quiet hours, delivery worker, and Notification Settings UI now coordinate real-time community alerts across achievements, streak milestones, agentic confirmations, and challenge feed events.
Read release notesEnabled multi-image uploads for the Labs Image→Quest pilot with combined OCR processing, per-image telemetry, and weekly cost tracking surfaced on admin dashboards.
Read release notesAutomated the 24-hour staging soak through the jobs runner with cron scheduling, artifact reports, and Slack alerts so Ops get daily Labs coverage without manual kickoff.
Read release notesAdded a Redis-backed memory backlog manager with telemetry so recall injections prioritise fresh entries and expose actionable triage IDs.
Read release notesHooked admin activation dashboards into the unified analytics ledger with install-source filters and CSV/PDF exports so ops and partners can audit retention without manual SQL.
Read release notesFinalized Sprint 25 infrastructure hardening: dedicated Upstash Redis topology with documented load test, failover drill, and post-cutover soak evidence ahead of RC readiness.
Read release notesHardened `/api/ops/performance-alert` validation so malformed payloads return typed 400 responses and Ops dispatch failures surface as 503 JSON responses.
Read release notesReplaces the Trials due-date picker with a responsive shadcn popover calendar so mobile users get a full-width, accessible control without triggering the keyboard.
Read release notesKeeps the Labs Image→Quest pilot safely gated for staged cohorts while preserving telemetry.
Read release notesRemoved legacy `invite_codes` table from gated beta phase (replaced by referral system in Sprint 17+)
Read release notesEnhanced personalization RAG suggestions with structured id, title, copy, and source fields.
Read release notesAdded a release-notes page at /help/release-notes with structured JSON data loading.
Read release notesCommunity feed refinements
Read release notesAdded a Sprint 24 completion report detailing soak outcomes, telemetry validation, and Labs readiness so Milestone 5 can close cleanly.
Read release notes