69 KiB
69 KiB
[APP NAME TBD] — Complete Product Blueprint
Science-Backed Manifestation & Thought Transformation App
Version: 1.0 — February 2026 Author: Matt (concept & research) + Claude (architecture & strategy) Status: Pre-development — Ready for implementation
Table of Contents
- Executive Summary
- The Problem & Opportunity
- Scientific Foundation
- The 6 Steps to Manifestation
- Product Vision & Core Features
- User Experience Flows
- AI Strategy & Prompt Architecture
- Technical Architecture
- Database Schema
- Monetization Model
- Cost Estimates
- Implementation Roadmap
- Building with Claude Code & Codex
- App Store Strategy
- Growth & Future Features
- Naming Candidates
- Research References
- Appendix: AI Prompt Templates
1. Executive Summary
This app is a science-backed personal transformation tool that does two things no competitor does well together:
- Transforms negative thoughts on the spot using AI personalized to the user's preferred coaching style — brutal honesty, gentle guidance, logical analysis, philosophical perspective, or humor.
- Guides users through a structured 6-step manifestation system grounded in peer-reviewed research from goal-setting theory, implementation intentions, self-efficacy, selective attention, habit formation, and network effects.
Unlike the vast majority of manifestation apps (which lean on "Law of Attraction" magical thinking), this app is built on published, peer-reviewed psychology and neuroscience research. The critical differentiator is Step 3: "Believe it's possible — but not guaranteed." This aligns with Bandura's self-efficacy theory (capability belief, not outcome certainty) and avoids the toxic positivity trap that causes most people to abandon manifestation when results aren't immediate.
Target users: Adults 18-45 interested in personal growth, goal achievement, and mental wellness — but skeptical of woo-woo self-help. The "I want to improve my life, but I want science, not crystals" market.
Revenue model: Freemium with premium subscription ($8/mo or $60/yr).
Tech stack: React Native + Expo (frontend) → Supabase (backend/database/auth) → Claude API (AI engine).
2. The Problem & Opportunity
The Problem
The self-help/manifestation app space is dominated by two extremes:
-
Pure woo-woo apps — Vision boards, affirmations, "raise your vibration" content. No scientific backing. Users get a dopamine hit, then quit when nothing changes because the app never connected belief to action.
-
Clinical CBT/therapy apps — Evidence-based but clinical, cold, and expensive. Feel like homework. Users who aren't in acute distress don't stick with them.
There's a massive gap in the middle: people who want to actively shape their lives using proven methods, delivered in a way that feels personal, warm, and actionable — not clinical or magical.
The Opportunity
- The meditation/mindfulness app market (Calm, Headspace) proved people will pay monthly for daily mental wellness tools. But meditation isn't for everyone — many people want active thought work, not passive stillness.
- AI makes personalized coaching scalable for the first time. What previously required a $200/hour therapist or coach (personalized thought reframing, goal scaffolding) can now happen instantly on a phone.
- The "manifestation" keyword has massive search volume but almost zero credible competition. Most results are embarrassingly unscientific. An app that owns "science-backed manifestation" has a blue ocean.
Competitive Landscape
| App | Category | Weakness vs. Our App |
|---|---|---|
| Reframe (alcohol reduction) | Behavioral change | Focused solely on drinking, not general thought transformation |
| Reframe AI (mental health) | CBT-based reframing | Generic AI coaches, no manifestation/goal system |
| Headspace / Calm | Meditation | Passive, not active thought work |
| Manifest (various) | Affirmations/vision boards | Zero science, pure "Law of Attraction" |
| BetterHelp / Talkspace | Therapy | Expensive ($240+/mo), requires scheduling, clinical positioning |
Our moat: The combination of (1) AI-powered personalized reframing with (2) a structured, research-cited goal achievement system, delivered with (3) anti-toxic-positivity guardrails and transparent science.
3. Scientific Foundation
Every feature in this app maps to published, peer-reviewed research. This isn't decoration — it's the product's structural integrity.
Research Pillars
Goal Setting & Clarity
- Locke & Latham (2002) — Building a Practically Useful Theory of Goal Setting and Task Motivation: A 35-Year Odyssey
- Locke, Latham, et al. — New Directions in Goal-Setting Theory
- Core insight: Specific, challenging goals with clear commitment lead to higher performance than vague "do your best" goals. This drives our Step 1 (Decide/Clarity) and the AI-guided SMART goal creation process.
Implementation Intentions (Planning & Action)
- Gollwitzer (1999) — Implementation Intentions: Strong Effects of Simple Plans
- Gollwitzer (2006) — Implementation Intentions and Goal Attainment
- Core insight: "If-then" plans ("If situation X arises, then I will do Y") dramatically increase follow-through on goals. This is the backbone of our Step 5 (Act in Alignment) and every micro-action the AI generates.
Visualization & Mental Practice
- Schuster et al. (2011) — Best Practice for Motor Imagery: A Systematic Literature Review
- Liu et al. (2025) — Effects of Imagery Practice on Athletes' Performance
- Seok et al. (2023) — Mental Practice in Stroke Rehabilitation
- Core insight: Mental rehearsal improves real-world performance across domains. Drives our Step 2 (See It) with personalized visualization scripts.
Self-Efficacy (Belief → Behavior)
- Bandura (1977) — Self-Efficacy: Toward a Unifying Theory of Behavioral Change
- Core insight: Belief in one's capability (not guaranteed outcomes) is the strongest predictor of goal achievement. This shapes our Step 3 (Believe It's Possible — But Not Guaranteed), which is the app's philosophical soul. We explicitly reject certainty-based belief ("the universe will provide") in favor of capability-based belief ("I have or can develop the skills to make this happen").
Attention & Selective Processing
- Yantis (2008) — The Neural Basis of Selective Attention
- Stevens & Bavelier (2012) — Attention, Learning, and Brain Plasticity
- Koch & Tsuchiya (2007) — Attention vs Consciousness
- Core insight: What you pay attention to shapes your reality — not metaphysically, but neurologically. Your brain filters millions of inputs daily, and attention training can literally change what you perceive. This drives our Step 4 (Notice Differently) and the Reframer's pattern analysis.
Habits & Repetition
- Wood, Mazar & Neal (2021) — Habits and Goals in Human Behavior
- Wood & Neal (2007) — A New Look at Habits and the Habit-Goal Interface
- Core insight: ~43% of daily behavior is habitual, not deliberate. Sustained change requires building automatic routines, not just willpower. Drives our Step 6 (Repeat and Compound) and streak tracking.
Network Effects & Opportunity
- Granovetter (1973) — The Strength of Weak Ties
- Core insight: Transformative opportunities come disproportionately from weak social ties, not close friends. Drives future community features where anonymous sharing creates unexpected connections and inspiration.
Expectancy Effects & Placebo
- Pardo-Cabello et al. (2022) — Placebo/Nocebo Effects: Mechanisms & Evidence
- Stetler et al. (2013) — Expectations & Placebo Response
- Core insight: Expectations directly influence physiological and behavioral outcomes. Supports the overall premise that structured positive expectation (when combined with action) measurably improves outcomes.
4. The 6 Steps to Manifestation
These steps form the app's conceptual spine. Every feature maps back to them.
┌─────────────────────────────────────────────────────────┐
│ │
│ Step 1: DECIDE (Clarity) │
│ ───────────────────────── │
│ Define exactly what you want and why it matters. │
│ Research: Locke & Latham goal-setting theory │
│ │
│ Step 2: SEE IT (Mental Rehearsal) │
│ ───────────────────────────────── │
│ Visualize the process and outcome with specificity. │
│ Research: Schuster et al., Liu et al. motor imagery │
│ │
│ Step 3: BELIEVE IT'S POSSIBLE (But Not Guaranteed) │
│ ────────────────────────────────────────────────── │
│ Build capability-based confidence, not blind faith. │
│ Research: Bandura self-efficacy theory │
│ │
│ Step 4: NOTICE DIFFERENTLY (Attention) │
│ ────────────────────────────────────── │
│ Train your brain to filter for aligned evidence. │
│ Research: Yantis, Stevens & Bavelier selective attn │
│ │
│ Step 5: ACT IN ALIGNMENT (Daily) │
│ ──────────────────────────────── │
│ Execute micro-actions using if-then planning. │
│ Research: Gollwitzer implementation intentions │
│ │
│ Step 6: REPEAT AND COMPOUND │
│ ─────────────────────────── │
│ Build automatic habits that sustain momentum. │
│ Research: Wood & Neal habit formation │
│ │
└─────────────────────────────────────────────────────────┘
Why Step 3 Is the App's Soul
Most manifestation content says: "Believe with absolute certainty and the universe will deliver."
This app says: "Believe you're capable of making this happen — but understand it's not guaranteed, and that's okay. Certainty isn't the point. Aligned action is."
This single distinction:
- Prevents the toxic positivity trap (and subsequent crash when things don't work)
- Aligns with Bandura's actual research (self-efficacy is about capability, not outcomes)
- Builds resilience when setbacks happen (because they were expected as possible)
- Makes the app credible to the skeptical-but-curious demographic that most manifestation apps repel
5. Product Vision & Core Features
The app has two core pillars that work independently but compound together.
Pillar 1: The Reframer (Daily Thought Transformer)
This is the app's sticky daily-use feature and primary acquisition tool. It's what users come for before they even know about the Manifestation Engine.
What it does: User inputs a negative thought or experience → AI generates a personalized reframe in their chosen coaching style → ends with a micro-action (if-then implementation intention).
Reframe Styles (user selects primary + secondary during onboarding):
| Style | Tone | Example Response |
|---|---|---|
| Brutal Honesty | Direct, no-nonsense | "You're catastrophizing. Here's what's actually true..." |
| Gentle Coach | Warm, empathetic | "That sounds really hard. Let's find what you can take from this..." |
| Logical Analyst | Structured, evidence-based | "Let's break this down. What evidence supports vs contradicts this thought?" |
| Philosophical | Big-picture, perspective-shifting | "Consider: what would this look like from 10 years in the future?" |
| Humor | Light, disarming | "Okay, worst case scenario... you become a cautionary tale in a self-help book. Best case?" |
Reframe Structure (every response follows this pattern):
- Acknowledge — Never dismiss the feeling. Validate first.
- Reframe — Apply the research-backed perspective shift. When appropriate, cite the science: "Your brain is doing what Yantis's research on selective attention predicts — you're filtering for the negative."
- Micro-action — End with a Gollwitzer-style if-then intention: "If [situation] happens again, then I will [specific response]."
Sub-features:
- Daily prompt — Push notification at user-set time: "What's weighing on you today?"
- Reframe history & pattern analysis — Over time, the app surfaces patterns: "You tend to catastrophize about work on Mondays" or "Your negative thoughts cluster around relationships." This is Step 4 (Notice Differently) applied to the user's own thought life.
- "Flip It" quick mode — One-tap mode for acute stress. Voice-record or quick-type → instant reframe. Zero friction.
Pillar 2: The Manifestation Engine (Goal System)
This is the premium feature. A structured journey through all 6 steps, guided by AI.
Step 1 — Decide (Clarity):
- AI guides user through Locke & Latham's SMART framework via conversational prompts
- Makes it feel personal, not corporate ("What would change in your life if you achieved this?" not "Define a measurable objective")
- Output: Crystal-clear goal statement + personal "why it matters" statement
- Free tier: Users can create ONE goal (enough to get invested, creates cliffhanger for premium)
Step 2 — See It (Mental Rehearsal):
- AI generates a personalized visualization script based on the user's goal
- Text format initially; optional audio in later phases (text-to-speech)
- Based on Schuster et al.'s motor imagery best practices: first-person, sensory-rich, process-focused (not just outcome)
- User sees this as a "Vision Summary" they can revisit daily
Step 3 — Believe It's Possible:
- AI walks user through identifying self-efficacy barriers
- Prompts: "What makes you doubt this is possible?" → systematically addresses each doubt
- Sources of evidence: past successes, transferable skills, role models, small wins
- Generates a "Belief Statement" — a personalized, evidence-based declaration of capability
- Anti-toxic-positivity: Never says "you definitely will." Says "you have real reasons to believe you can."
Step 4 — Notice Differently:
- Daily "attention prompts" — "Today, notice one thing that's already aligned with your goal"
- Trains selective attention (Yantis, Stevens & Bavelier) toward goal-aligned evidence
- Users log observations → builds an "Evidence Journal" that reinforces belief over time
- Over time, AI surfaces patterns: "You've noticed 23 instances of alignment this month. Your brain is rewiring."
Step 5 — Act in Alignment:
- AI generates daily micro-actions using Gollwitzer's if-then structure
- Format: "If it's [time/situation], then I will [specific 15-minute action toward goal]"
- Users check off completed actions
- Streak tracking with smart reminders
- AI adapts difficulty: starts tiny (2-minute actions), scales up as habit solidifies
Step 6 — Repeat & Compound:
- Habit tracking dashboard showing compound progress over time
- Visual growth charts, milestone celebrations
- AI periodically revisits and updates the plan based on progress
- Weekly AI-generated summary: reframes, patterns, goal progress, suggested adjustments
- Celebrates the process (consistency) not just outcomes
Additional Features
Anti-Manifestation Guard: When the AI detects a user spiraling into toxic positivity, magical thinking, or unhealthy attachment to outcomes, it gently course-corrects:
"Manifestation isn't about wishing — it's about aligning your attention, beliefs, and actions with what you want. Let's focus on what you can control today."
Science Drawer: Each step has an expandable "Why This Works" section citing the actual research. Example:
"This technique is based on Gollwitzer's 1999 research on Implementation Intentions, which showed that people who create if-then plans are 2-3x more likely to follow through on goals than those who simply set intentions."
This builds trust and differentiates from every competitor that says "trust the process" without explaining why.
Weekly Reflection: AI-generated weekly summary covering:
- Reframe count and patterns
- Mood trajectory (inferred from thought inputs)
- Goal progress against daily actions
- Suggested adjustments for the coming week
Streak & Gamification (Light):
- Daily streaks for reframing and goal actions
- "Reframe count" milestone celebrations (10, 50, 100, 500...)
- Level-ups tied to consistency (not volume — doing it daily matters more than doing 10 in one day)
- No leaderboards, no competitive mechanics. This is personal transformation, not a game.
6. User Experience Flows
Flow 1: First Open (Onboarding)
1. Welcome screen
"Transform your thinking. Achieve your goals. Backed by science."
[Get Started]
2. How this works (3 swipeable cards)
Card 1: "Reframe your negative thoughts daily with AI"
Card 2: "Build a science-backed plan for what you want in life"
Card 3: "Watch your thinking patterns transform over time"
3. Choose your reframe style
"When life gets hard, how do you prefer to hear the truth?"
[Show 5 style cards with examples — user taps primary, then secondary]
4. Set your daily check-in time
"When should we ask how you're doing?"
[Time picker — defaults to 8:00 PM]
5. Create account
[Email / Google / Apple Sign-In]
6. First reframe
"Let's try it right now. What's been on your mind?"
[Text input → AI reframe → "wow" moment]
7. Home screen
Flow 2: Daily Reframe
1. Push notification: "What's weighing on you today?"
[User taps → opens app to input screen]
2. Input screen
Text field: "What happened? What's bothering you?"
[Optional: voice input button]
[Submit]
3. Processing (1-3 seconds)
Brief animation (not a spinner — something warm)
4. Reframe card
┌─────────────────────────────────────┐
│ Your thought: │
│ "I bombed the presentation and │
│ everyone thinks I'm incompetent" │
│ │
│ [Brutal Honesty] reframe: │
│ │
│ "Let's be real: you're not a mind │
│ reader. You have zero evidence that │
│ 'everyone' thinks anything. What │
│ you DO know is one presentation │
│ didn't go as planned..." │
│ │
│ 💡 Your micro-action: │
│ "If I start replaying the │
│ presentation, then I will write │
│ down 3 things that DID go right." │
│ │
│ 📚 Why this works ▼ │
│ │
│ [Save] [Share] [Try Different Style] │
└─────────────────────────────────────┘
5. Save → returns to home with updated streak
Flow 3: Manifestation Engine (Goal Creation Wizard)
1. "Create a Manifestation" [+]
2. Step 1: Decide
AI conversation (3-5 exchanges):
"What do you want to achieve?"
→ "Why does this matter to you personally?"
→ "What would your life look like if this happened?"
→ "Let's make this specific..." [SMART refinement]
→ Output: Clarity Statement displayed as a card
3. Step 2: See It
AI generates visualization script
"Close your eyes and imagine..."
[User reads/listens → marks as complete]
4. Step 3: Believe
AI conversation:
"What makes you doubt this is possible?"
→ [User lists doubts]
→ AI addresses each with evidence + past successes
→ Output: Belief Statement card
5. Step 4: Notice
AI sets up attention prompts
"Starting tomorrow, I'll send you one prompt per day"
[Preview first prompt]
6. Step 5: Act
AI generates first week of if-then micro-actions
[User reviews and confirms]
7. Step 6: Compound
Dashboard appears showing the full system:
- Goal card
- Today's attention prompt
- Today's micro-action
- Streak counter
- Progress timeline
8. "Your manifestation engine is running. Show up daily."
Flow 4: Weekly Summary
Push notification (Sunday evening):
"Your weekly transformation report is ready ✨"
┌─────────────────────────────────────┐
│ WEEK 3 SUMMARY │
│ │
│ 🔄 Reframes this week: 6 │
│ Pattern: Work stress (Mon/Tue) │
│ Shift: Catastrophizing → present │
│ │
│ 🎯 Goal: [user's manifestation] │
│ Actions completed: 5/7 │
│ Evidence logged: 4 observations │
│ │
│ 📈 Trend: Your negative thoughts │
│ are 23% less catastrophic than │
│ Week 1. Your brain is learning. │
│ │
│ 💡 Suggestion for next week: │
│ "Your Tuesday stress pattern │
│ suggests a preparation ritual │
│ Monday night could help..." │
│ │
│ [View Full Report] [Share Progress] │
└─────────────────────────────────────┘
7. AI Strategy & Prompt Architecture
Why Claude API
Claude is the right AI engine for this app because:
- Tone control — Claude excels at adopting specific communication styles while maintaining safety guardrails
- Nuance — Thought reframing requires acknowledging complexity, not defaulting to toxic positivity. Claude handles nuance better than most models.
- Safety — Claude's constitutional AI approach means it won't encourage harmful thinking patterns, which is critical for a mental wellness app
- Cost — Claude Haiku (for simple reframes) and Sonnet (for complex goal guidance) provide a cost-effective tier structure
Model Selection by Feature
| Feature | Model | Rationale | Est. Cost/Call |
|---|---|---|---|
| Simple reframe | Claude Haiku 4.5 | Fast, cheap, good enough for single-thought reframes | ~$0.003 |
| Complex reframe (with patterns) | Claude Sonnet 4.5 | Needs to analyze history + generate nuanced response | ~$0.008 |
| Goal creation wizard | Claude Sonnet 4.5 | Multi-turn conversation requiring coherent guidance | ~$0.015 |
| Visualization script | Claude Sonnet 4.5 | Creative, sensory-rich, personalized writing | ~$0.012 |
| Weekly summary | Claude Sonnet 4.5 | Pattern analysis across multiple data points | ~$0.020 |
| Pattern detection | Claude Haiku 4.5 | Categorization task, doesn't need deep reasoning | ~$0.003 |
System Prompt Architecture
All AI calls go through Supabase Edge Functions (never direct from client). Each function includes:
- System prompt — Defines the AI's role, constraints, and output format
- User context — Injected from the database (reframe style, goal data, history)
- Research grounding — Relevant citations pre-loaded so the AI can reference them naturally
- Anti-toxicity guardrails — Explicit instructions to avoid toxic positivity, magical thinking, and dismissal of real problems
See Appendix: AI Prompt Templates for full prompt examples.
Rate Limiting & Cost Control
- Free tier: 2 reframes/day (hard limit enforced at Edge Function level)
- Premium tier: Unlimited (soft limit of 20/day with warning, hard limit of 50/day)
- All responses cached for 24 hours — if user submits identical thought, return cached reframe
- Goal creation wizard: Max 10 AI exchanges per goal setup
- Weekly summaries: Generated once per week via cron job, not on-demand
8. Technical Architecture
Stack Overview
| Layer | Technology | Why This Choice |
|---|---|---|
| Frontend | React Native + Expo | Single codebase for iOS, Android, web. JavaScript/TypeScript (most transferable skill). Expo handles 90% of native complexity. Huge community + AI coding tool support. |
| Backend/DB | Supabase | Managed Postgres DB, built-in auth (email, Google, Apple Sign-In), Edge Functions (Deno runtime), file storage, real-time subscriptions. No servers to manage. |
| AI Engine | Claude API (Anthropic) | Called via Supabase Edge Functions. Never direct from client (cost control, rate limiting, prompt security). |
| Push Notifications | Expo Push Notifications | Free, built into Expo, works on both platforms out of the box. |
| Payments | RevenueCat | Handles App Store + Google Play subscriptions. Way easier than native in-app purchase implementation. Manages receipts, trials, and entitlements. |
| App Builds | Expo EAS | Cloud builds for iOS and Android. Handles code signing, certificates, TestFlight, and OTA updates. |
| Analytics | PostHog (or Mixpanel free tier) | Open-source product analytics. Event tracking, funnels, retention analysis. |
Architecture Diagram
┌──────────────────────────────────────────────────────────────┐
│ User's Phone / Web Browser │
│ (React Native + Expo App) │
│ │
│ ┌─────────┐ ┌──────────┐ ┌──────────┐ ┌────────────────┐ │
│ │Reframer │ │Manifest. │ │ History │ │ Settings │ │
│ │ Screen │ │ Engine │ │ & Stats │ │ & Profile │ │
│ └────┬────┘ └────┬─────┘ └────┬─────┘ └───────┬────────┘ │
│ │ │ │ │ │
│ └───────────┴────────────┴────────────────┘ │
│ │ │
│ Supabase JS Client SDK │
└──────────────────────────┬───────────────────────────────────┘
│ HTTPS
▼
┌──────────────────────────────────────────────────────────────┐
│ Supabase │
│ │
│ ┌──────────────────────────────────────────────────────┐ │
│ │ Auth (email, Google Sign-In, Apple Sign-In) │ │
│ │ - JWT tokens, session management │ │
│ │ - Row Level Security (RLS) on all tables │ │
│ └──────────────────────────────────────────────────────┘ │
│ │
│ ┌──────────────────────────────────────────────────────┐ │
│ │ PostgreSQL Database │ │
│ │ - profiles (user preferences, reframe style) │ │
│ │ - reframes (thought log + AI responses) │ │
│ │ - manifestations (goals, clarity statements) │ │
│ │ - daily_actions (if-then micro-actions) │ │
│ │ - attention_logs (evidence journal) │ │
│ │ - streaks (habit tracking) │ │
│ │ - subscriptions (premium status via RevenueCat) │ │
│ └──────────────────────────────────────────────────────┘ │
│ │
│ ┌──────────────────────────────────────────────────────┐ │
│ │ Edge Functions (Deno TypeScript) │ │
│ │ - POST /reframe → validates, calls Claude, saves │ │
│ │ - POST /generate-plan → goal wizard AI calls │ │
│ │ - POST /generate-visualization → Step 2 scripts │ │
│ │ - POST /weekly-summary → pattern analysis + report │ │
│ │ - POST /check-subscription → RevenueCat validation │ │
│ │ - All functions enforce rate limits & auth │ │
│ └──────────────────────────────────────────────────────┘ │
│ │
│ ┌──────────────────────────────────────────────────────┐ │
│ │ Storage │ │
│ │ - Profile pictures │ │
│ │ - Audio visualization scripts (Phase 2+) │ │
│ │ - Exported reports (PDF generation, Phase 3+) │ │
│ └──────────────────────────────────────────────────────┘ │
│ │
│ ┌──────────────────────────────────────────────────────┐ │
│ │ Cron Jobs (pg_cron or Supabase scheduled functions) │ │
│ │ - Weekly summary generation (Sundays 6pm user TZ) │ │
│ │ - Streak reset check (daily midnight user TZ) │ │
│ │ - Pattern analysis batch job (weekly) │ │
│ └──────────────────────────────────────────────────────┘ │
└──────────────────────────┬───────────────────────────────────┘
│ HTTPS (from Edge Functions only)
▼
┌──────────────────────────────────────────────────────────────┐
│ Claude API (Anthropic) │
│ │
│ - Thought reframing (Haiku for simple, Sonnet for complex) │
│ - Goal guidance & conversational wizards │
│ - Visualization script generation │
│ - Pattern analysis & weekly summaries │
│ - Anti-toxicity guardrails baked into system prompts │
└──────────────────────────────────────────────────────────────┘
┌──────────────────────────────────────────────────────────────┐
│ External Services │
│ │
│ RevenueCat ──── Subscription management, receipt validation │
│ Expo EAS ────── Cloud builds, OTA updates, push notifs │
│ PostHog ─────── Product analytics, funnel tracking │
└──────────────────────────────────────────────────────────────┘
Key Architecture Decisions
Why Supabase Edge Functions for AI calls (not direct from client):
- API key security — never expose the Claude API key in the client bundle
- Rate limiting — enforce free/premium limits server-side
- Prompt security — system prompts stay on the server (can't be extracted)
- Cost control — can add caching, deduplication, and usage tracking
- Response formatting — can sanitize/validate AI responses before sending to client
Why Expo over bare React Native:
- Managed workflow — no need to touch Xcode or Android Studio for most tasks
- EAS Build — cloud builds mean Matt doesn't need a Mac for iOS builds
- OTA updates — push bug fixes without App Store review
- Built-in push notifications — no Firebase setup needed
- Web support — same codebase runs on web via Expo Web
Why RevenueCat over native IAP:
- Handles the nightmare of App Store receipt validation
- Single API for both iOS and Android subscriptions
- Built-in analytics (MRR, churn, trial conversion)
- Webhook support for server-side subscription status
- Free up to $2,500/mo MTR
9. Database Schema
Core Tables
-- ==========================================
-- PROFILES
-- User preferences and settings
-- ==========================================
CREATE TABLE profiles (
id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
user_id UUID REFERENCES auth.users(id) ON DELETE CASCADE NOT NULL UNIQUE,
display_name TEXT,
reframe_style_primary TEXT NOT NULL DEFAULT 'gentle_coach'
CHECK (reframe_style_primary IN (
'brutal_honesty', 'gentle_coach', 'logical_analyst',
'philosophical', 'humor'
)),
reframe_style_secondary TEXT
CHECK (reframe_style_secondary IN (
'brutal_honesty', 'gentle_coach', 'logical_analyst',
'philosophical', 'humor'
)),
notification_time TIME DEFAULT '20:00:00',
timezone TEXT DEFAULT 'UTC',
onboarding_completed BOOLEAN DEFAULT FALSE,
subscription_tier TEXT DEFAULT 'free'
CHECK (subscription_tier IN ('free', 'premium')),
reframes_today INTEGER DEFAULT 0,
last_reframe_date DATE,
created_at TIMESTAMPTZ DEFAULT NOW(),
updated_at TIMESTAMPTZ DEFAULT NOW()
);
-- ==========================================
-- REFRAMES
-- Thought transformation history
-- ==========================================
CREATE TABLE reframes (
id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
user_id UUID REFERENCES auth.users(id) ON DELETE CASCADE NOT NULL,
original_thought TEXT NOT NULL,
reframed_response TEXT NOT NULL,
reframe_style_used TEXT NOT NULL,
category TEXT, -- AI-inferred: 'work', 'relationships', 'health', 'finance', 'self-worth', 'general'
sentiment_score DECIMAL(3,2), -- -1.00 to 1.00 (negative to positive)
if_then_action TEXT, -- The micro-action from the reframe
is_quick_flip BOOLEAN DEFAULT FALSE, -- Was this a "Flip It" quick reframe?
created_at TIMESTAMPTZ DEFAULT NOW()
);
-- Index for history queries and pattern analysis
CREATE INDEX idx_reframes_user_date ON reframes(user_id, created_at DESC);
CREATE INDEX idx_reframes_user_category ON reframes(user_id, category);
-- ==========================================
-- MANIFESTATIONS
-- Goal/manifestation tracking (the 6-step engine)
-- ==========================================
CREATE TABLE manifestations (
id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
user_id UUID REFERENCES auth.users(id) ON DELETE CASCADE NOT NULL,
title TEXT NOT NULL,
-- Step 1: Decide
clarity_statement TEXT,
why_it_matters TEXT,
-- Step 2: See It
visualization_text TEXT,
visualization_audio_url TEXT, -- Future: TTS audio file
-- Step 3: Believe
belief_statement TEXT,
doubts_addressed JSONB, -- Array of {doubt, response} objects
-- Metadata
status TEXT DEFAULT 'active'
CHECK (status IN ('active', 'paused', 'completed', 'archived')),
current_step INTEGER DEFAULT 1 CHECK (current_step BETWEEN 1 AND 6),
created_at TIMESTAMPTZ DEFAULT NOW(),
updated_at TIMESTAMPTZ DEFAULT NOW()
);
CREATE INDEX idx_manifestations_user ON manifestations(user_id, status);
-- ==========================================
-- DAILY ACTIONS
-- If-then micro-actions (Step 5)
-- ==========================================
CREATE TABLE daily_actions (
id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
user_id UUID REFERENCES auth.users(id) ON DELETE CASCADE NOT NULL,
manifestation_id UUID REFERENCES manifestations(id) ON DELETE CASCADE NOT NULL,
if_trigger TEXT NOT NULL, -- "If it's 7am..."
then_action TEXT NOT NULL, -- "...then I will spend 15 min writing"
scheduled_date DATE NOT NULL,
completed BOOLEAN DEFAULT FALSE,
completed_at TIMESTAMPTZ,
created_at TIMESTAMPTZ DEFAULT NOW()
);
CREATE INDEX idx_daily_actions_user_date ON daily_actions(user_id, scheduled_date);
-- ==========================================
-- ATTENTION LOGS
-- Evidence journal entries (Step 4)
-- ==========================================
CREATE TABLE attention_logs (
id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
user_id UUID REFERENCES auth.users(id) ON DELETE CASCADE NOT NULL,
manifestation_id UUID REFERENCES manifestations(id) ON DELETE CASCADE NOT NULL,
prompt_text TEXT, -- The attention prompt that was shown
observation TEXT NOT NULL, -- What the user noticed
created_at TIMESTAMPTZ DEFAULT NOW()
);
CREATE INDEX idx_attention_logs_user ON attention_logs(user_id, created_at DESC);
-- ==========================================
-- STREAKS
-- Habit tracking data
-- ==========================================
CREATE TABLE streaks (
id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
user_id UUID REFERENCES auth.users(id) ON DELETE CASCADE NOT NULL,
streak_type TEXT NOT NULL
CHECK (streak_type IN ('reframe', 'daily_action', 'attention_log', 'overall')),
current_count INTEGER DEFAULT 0,
longest_count INTEGER DEFAULT 0,
last_activity_date DATE,
created_at TIMESTAMPTZ DEFAULT NOW(),
updated_at TIMESTAMPTZ DEFAULT NOW(),
UNIQUE(user_id, streak_type)
);
-- ==========================================
-- WEEKLY SUMMARIES
-- Pre-generated weekly reports
-- ==========================================
CREATE TABLE weekly_summaries (
id UUID PRIMARY KEY DEFAULT gen_random_uuid(),
user_id UUID REFERENCES auth.users(id) ON DELETE CASCADE NOT NULL,
week_start DATE NOT NULL,
week_end DATE NOT NULL,
summary_text TEXT NOT NULL,
reframe_count INTEGER DEFAULT 0,
action_completion_rate DECIMAL(5,2),
attention_log_count INTEGER DEFAULT 0,
top_category TEXT,
mood_trend TEXT, -- 'improving', 'stable', 'declining'
created_at TIMESTAMPTZ DEFAULT NOW()
);
CREATE INDEX idx_weekly_summaries_user ON weekly_summaries(user_id, week_start DESC);
-- ==========================================
-- ROW LEVEL SECURITY
-- Users can only access their own data
-- ==========================================
ALTER TABLE profiles ENABLE ROW LEVEL SECURITY;
ALTER TABLE reframes ENABLE ROW LEVEL SECURITY;
ALTER TABLE manifestations ENABLE ROW LEVEL SECURITY;
ALTER TABLE daily_actions ENABLE ROW LEVEL SECURITY;
ALTER TABLE attention_logs ENABLE ROW LEVEL SECURITY;
ALTER TABLE streaks ENABLE ROW LEVEL SECURITY;
ALTER TABLE weekly_summaries ENABLE ROW LEVEL SECURITY;
-- Policy template (apply to each table):
CREATE POLICY "Users can only access own data" ON profiles
FOR ALL USING (auth.uid() = user_id);
-- Repeat for each table...
Schema Relationships
auth.users (Supabase built-in)
│
├── profiles (1:1)
│
├── reframes (1:many)
│
├── manifestations (1:many)
│ │
│ ├── daily_actions (1:many per manifestation)
│ │
│ └── attention_logs (1:many per manifestation)
│
├── streaks (1:4 — one per streak_type)
│
└── weekly_summaries (1:many — one per week)
10. Monetization Model
Freemium with Smart Gate
Free Tier (acquisition & habit building):
- 2 reframes per day (enough to build habit, not enough to feel unlimited)
- Access to Step 1 of Manifestation Engine (Decide/Clarity) — user creates a goal and feels invested, but can't access the full system
- Reframe history (last 7 days only)
- Basic streak tracking
- All research citations accessible
Premium ($8/month or $60/year):
- Unlimited reframes (soft limit 20/day, hard limit 50/day)
- Full 6-step Manifestation Engine
- Complete reframe history + pattern analysis
- Weekly AI summaries
- Multiple active manifestations (up to 5 concurrent goals)
- Audio visualizations (when available)
- Priority AI response (Sonnet for all calls vs. Haiku for free tier)
Pricing Strategy
- $8/month — The "coffee a week" mental anchor. Positioned against therapy ($200+/session) and coaching ($100+/session).
- $60/year ($5/month effective) — 25% discount incentivizes annual commitment. Display prominently as the recommended option.
- 7-day free trial for premium features — Lets users experience the full Manifestation Engine before committing.
- No lifetime purchase option initially — recurring revenue is more sustainable and aligns with the "daily practice" message.
Revenue Projections (Conservative)
| Metric | Month 3 | Month 6 | Month 12 |
|---|---|---|---|
| Total downloads | 1,000 | 5,000 | 20,000 |
| Free → Premium conversion | 5% | 5% | 5% |
| Paying users | 50 | 250 | 1,000 |
| ARPU (blended monthly/annual) | $6.50 | $6.50 | $6.50 |
| MRR | $325 | $1,625 | $6,500 |
| ARR | $3,900 | $19,500 | $78,000 |
These are conservative estimates assuming organic growth + basic ASO. Paid acquisition or viral mechanics could significantly accelerate.
11. Cost Estimates
Launch Phase (0-1,000 users)
| Service | Plan | Cost |
|---|---|---|
| Supabase | Free tier (500MB DB, 50K auth, 500K edge invocations) | $0 |
| Claude API | Pay-as-you-go (~$0.003-0.015/call) | $50-100/mo |
| Expo EAS | Free tier (30 builds/month) | $0 |
| RevenueCat | Free (up to $2,500 MTR) | $0 |
| PostHog | Free tier (1M events/mo) | $0 |
| Apple Developer | Annual fee | $99/yr ($8.25/mo) |
| Google Play Developer | One-time | $25 |
| Total | ~$60-110/mo |
Growth Phase (1,000-10,000 users)
| Service | Plan | Cost |
|---|---|---|
| Supabase | Pro ($25/mo) | $25/mo |
| Claude API | Increased usage | $100-200/mo |
| Expo EAS | Production ($99/mo) | $99/mo |
| RevenueCat | 1% of revenue after $2.5K MTR | Variable |
| PostHog | Still free tier likely | $0 |
| Total | ~$225-325/mo |
Breakeven Analysis
At $6.50 ARPU, breakeven on $250/mo infrastructure requires ~39 paying users. With 5% conversion, that's ~780 total users. Achievable within the first 2-3 months with moderate marketing effort.
12. Implementation Roadmap
Phase 1 — MVP: The Reframer (4-6 weeks)
Goal: Ship the core reframing feature. Get it into users' hands. Validate the concept.
Week 1-2: Foundation
- Initialize Expo project with TypeScript
- Set up Supabase project (database, auth, edge functions)
- Implement auth flow (email + Google + Apple Sign-In)
- Create database tables and RLS policies
- Build basic navigation structure (tab navigator: Home, Reframe, History, Profile)
Week 2-3: Onboarding
- Welcome screens (swipeable intro cards)
- Reframe style selection (5 styles with example previews)
- Notification time picker
- Profile creation flow
Week 3-4: Core Reframer
- Text input screen for negative thoughts
- Supabase Edge Function:
/reframe— takes thought + user context, calls Claude, returns reframe - Reframe display card (with the three-part structure: acknowledge → reframe → micro-action)
- "Why This Works" expandable science section
- Save reframe to database
- "Try Different Style" button (re-call with different style)
Week 4-5: History & Streaks
- Reframe history list (chronological, filterable by category)
- Basic streak counter (current + longest)
- Daily reframe counter (enforce free tier limit)
- Push notification for daily check-in
Week 5-6: Polish & Internal Testing
- Error handling and loading states
- Offline support (queue reframe requests)
- App icon, splash screen, basic branding
- Internal TestFlight / Google Play internal testing
Phase 2 — Manifestation Engine (4-6 weeks)
Goal: Build the full 6-step goal system. This is the premium feature.
Week 7-8: Goal Creation Wizard (Steps 1-3)
- Conversational AI flow for Step 1 (Decide) — multi-turn chat interface
- Visualization script generation for Step 2 (See It)
- Self-efficacy barrier identification for Step 3 (Believe)
- Save manifestation with all generated content
Week 9-10: Daily System (Steps 4-5)
- Attention prompts (push notifications with goal-specific prompts)
- Evidence journal (log observations)
- Daily if-then action generation and tracking
- Check-off interface with streak integration
Week 10-11: Compound & Summary (Step 6)
- Progress dashboard (visual timeline of actions, evidence, streaks)
- Weekly AI summary generation (cron job + display)
- Milestone celebrations
Week 11-12: Subscription Gate
- RevenueCat integration (iOS + Android)
- Paywall screen (appears when free user tries to access Steps 2-6)
- Premium entitlement checks throughout the app
- 7-day free trial flow
- Restore purchases functionality
Phase 3 — Polish & Launch (2-4 weeks)
Week 13-14: App Store Preparation
- App Store screenshots and preview video
- App Store description (ASO-optimized)
- Privacy policy and terms of service
- App Store review compliance check
- Submit to Apple App Store and Google Play
Week 14-16: Launch & Iteration
- Analytics integration (PostHog events for all key actions)
- Crash reporting (Sentry or similar)
- Bug fixes from beta feedback
- Web version deployment (Expo Web → Vercel/Netlify)
- First marketing push
Phase 4 — Growth Features (Ongoing)
- Voice input for quick reframes (Expo Audio → Whisper API or native STT)
- Audio visualization scripts (TTS generation, downloadable)
- Community features (anonymous sharing of reframes and wins)
- Pattern analysis dashboard (deeper insights into thought patterns)
- Widget support (iOS/Android home screen widget showing streak + daily prompt)
- Dark mode
- Multi-language support
13. Building with Claude Code & Codex
This section provides the exact workflow for building the app using AI-assisted development tools.
Development Environment Setup
# 1. Install Node.js (LTS version)
# Download from https://nodejs.org or use nvm:
nvm install --lts
nvm use --lts
# 2. Install Expo CLI
npm install -g @expo/cli
# 3. Create the project
npx create-expo-app@latest [app-name] --template tabs
cd [app-name]
# 4. Install core dependencies
npx expo install @supabase/supabase-js
npx expo install expo-secure-store # For secure token storage
npx expo install expo-notifications # Push notifications
npx expo install react-native-purchases # RevenueCat
npx expo install expo-av # Audio playback (future)
# 5. Install navigation & UI
npx expo install @react-navigation/native @react-navigation/bottom-tabs
npx expo install react-native-screens react-native-safe-area-context
# 6. Development dependencies
npm install -D typescript @types/react
Claude Code Workflow
Claude Code (the CLI tool) is ideal for this project. Here's how to use it effectively:
Project initialization prompt:
I'm building a React Native + Expo app with Supabase backend.
The app is a science-backed manifestation tool with two core features:
1. AI-powered thought reframing (user inputs negative thought → Claude API reframes it)
2. A 6-step goal achievement system
Tech stack: Expo (TypeScript), Supabase (Postgres + Edge Functions + Auth), Claude API.
Monetization: Freemium with RevenueCat.
Start by setting up the project structure, Supabase client configuration,
and auth flow (email + Google + Apple Sign-In).
Iterative development approach:
- Describe the feature you want to build
- Let Claude Code scaffold the files
- Review and test
- Iterate with specific refinements
- Move to next feature
Recommended Claude Code commands for this project:
# Start Claude Code in the project directory
claude
# Example prompts for each phase:
# Auth setup
"Set up Supabase auth with email, Google, and Apple Sign-In.
Create a useAuth hook and an AuthContext provider."
# Reframer
"Create the reframe input screen with a text area, submit button,
and loading state. It should call a Supabase Edge Function at /reframe."
# Edge Function
"Write a Supabase Edge Function that takes a negative thought and user's
reframe style preference, calls the Claude API with a system prompt for
thought reframing, and returns the structured response."
# Database
"Generate the SQL migration for creating the reframes table with RLS policies."
Codex (OpenAI) Workflow
Codex can be used alongside Claude Code for:
- Generating boilerplate React Native components
- Writing database migration scripts
- Creating test files
- Generating TypeScript types from the schema
Complementary usage: Use Claude Code for architecture decisions and complex logic. Use Codex for repetitive scaffolding and boilerplate generation. They don't conflict — they complement.
File Structure
[app-name]/
├── app/ # Expo Router file-based routing
│ ├── (tabs)/ # Tab navigator screens
│ │ ├── index.tsx # Home / Dashboard
│ │ ├── reframe.tsx # Reframe input screen
│ │ ├── goals.tsx # Manifestation Engine
│ │ ├── history.tsx # Reframe history
│ │ └── profile.tsx # Settings & profile
│ ├── (auth)/ # Auth screens
│ │ ├── login.tsx
│ │ ├── register.tsx
│ │ └── onboarding.tsx
│ ├── reframe/
│ │ └── [id].tsx # Individual reframe detail
│ ├── manifestation/
│ │ ├── create.tsx # Goal creation wizard
│ │ └── [id].tsx # Goal detail / dashboard
│ └── _layout.tsx # Root layout
├── components/
│ ├── ui/ # Generic UI components
│ │ ├── Button.tsx
│ │ ├── Card.tsx
│ │ ├── TextInput.tsx
│ │ └── LoadingSpinner.tsx
│ ├── reframe/
│ │ ├── ReframeCard.tsx # Displays a reframe result
│ │ ├── StyleSelector.tsx # Reframe style picker
│ │ └── PatternInsight.tsx # Pattern analysis display
│ ├── manifestation/
│ │ ├── StepProgress.tsx # 6-step progress indicator
│ │ ├── ActionCard.tsx # Daily action card
│ │ └── EvidenceEntry.tsx # Attention log entry
│ └── shared/
│ ├── ScienceDrawer.tsx # "Why This Works" expandable
│ └── StreakBadge.tsx # Streak display component
├── hooks/
│ ├── useAuth.ts # Authentication hook
│ ├── useReframe.ts # Reframe CRUD operations
│ ├── useManifestation.ts # Goal CRUD operations
│ ├── useStreak.ts # Streak tracking
│ └── useSubscription.ts # RevenueCat integration
├── lib/
│ ├── supabase.ts # Supabase client init
│ ├── constants.ts # App-wide constants
│ └── types.ts # TypeScript interfaces
├── contexts/
│ ├── AuthContext.tsx # Auth state provider
│ └── SubscriptionContext.tsx # Premium status provider
├── supabase/
│ ├── migrations/ # SQL migration files
│ │ ├── 001_create_profiles.sql
│ │ ├── 002_create_reframes.sql
│ │ ├── 003_create_manifestations.sql
│ │ ├── 004_create_daily_actions.sql
│ │ ├── 005_create_attention_logs.sql
│ │ ├── 006_create_streaks.sql
│ │ └── 007_create_weekly_summaries.sql
│ └── functions/ # Edge Functions
│ ├── reframe/index.ts
│ ├── generate-plan/index.ts
│ ├── generate-visualization/index.ts
│ ├── weekly-summary/index.ts
│ └── check-subscription/index.ts
├── assets/
│ ├── images/
│ └── fonts/
├── app.json # Expo config
├── tsconfig.json
└── package.json
14. App Store Strategy
App Store Optimization (ASO)
Primary keywords to target:
- "manifestation app"
- "reframe negative thoughts"
- "positive thinking app"
- "goal setting app"
- "science-based self help"
- "thought reframing"
- "daily affirmation" (high volume, position as the "scientific" alternative)
App Store title format:
[App Name] — Science-Backed Manifestation
Subtitle (30 chars):
Reframe Thoughts. Achieve Goals.
Description strategy: Lead with the unique differentiator (science-backed), include social proof (research citations), address the skeptic ("not another woo-woo app"), and end with clear feature list.
Review Strategy
- Prompt for review after the user's 7th reframe (proven they find value)
- Never prompt during or after a negative reframe (bad timing)
- Prompt after completing Step 1 of a manifestation (feeling accomplished)
Privacy & Compliance
- All data encrypted at rest (Supabase handles this)
- GDPR compliance: data export and deletion capability
- No third-party data sharing
- Mental health disclaimer: "This app is not a substitute for professional therapy or medical advice"
- Terms of service clearly stating AI-generated content limitations
- Age gate: 17+ (due to AI-generated content about potentially sensitive thoughts)
15. Growth & Future Features
Phase 4+ Feature Ideas (Prioritized)
High Impact / Lower Effort:
- iOS/Android widget — Home screen widget showing streak count + "Reframe Now" button. Reduces friction to zero.
- Dark mode — Table stakes for a daily-use app.
- Voice input — Speak your thought instead of typing. Uses Expo Audio + speech-to-text.
- Share reframes — Export a reframe as a beautiful card image for social sharing (viral loop).
High Impact / Higher Effort: 5. Audio visualizations — Text-to-speech generation of visualization scripts. Users can listen during meditation or commute. 6. Community — Anonymous sharing of reframes and wins. Based on Granovetter's weak ties research. Users see how others transformed similar thoughts. 7. Pattern dashboard — Deep analytics on thought categories, time patterns, sentiment trends. "Your catastrophizing has decreased 40% since you started." 8. Partner/friend feature — Invite someone to share streaks and encourage each other (optional, privacy-first).
Moonshot / Future: 9. AI-generated daily podcast — Personalized 3-minute audio each morning based on current goals and recent patterns. 10. Integration with calendar — Auto-schedule if-then actions as calendar blocks. 11. Wearable integration — Apple Watch / WearOS app for quick-flip reframes. 12. Therapist dashboard — Optional feature where users can share their reframe/goal data with their actual therapist.
Potential Pivots / Expansions
- B2B: Corporate wellness programs (companies buy bulk subscriptions for employees)
- Education: Student-focused version for schools/universities
- Coaching platform: Let certified coaches use the app as a tool with their clients, with shared dashboards
16. Naming Candidates
The app name is TBD. Here are the current candidates explored and their status:
Taken / Conflicts
| Name | Issue |
|---|---|
| Reframe | Multiple apps (alcohol reduction, AI mental health, mindfulness) |
| MindShift | Existing CBT-based anxiety app |
| Prism | Multiple apps |
| Forge / MindForge | Various tech companies |
| Lumen | Metabolism tracker app |
| Myndset | Multiple companies (therapeutics, ideation) |
| Neurashift | AI consulting company |
| Mentavi | ADHD diagnostic platform |
| Soteria | Psychiatric treatment term |
| GrowMind | IT solutions + e-learning |
| Mindrift | AI freelance platform |
Current Shortlist (Availability Needs Final Verification)
Clareo
- Latin: "clareo" = to be bright, to be clear
- Pronunciation: "cla-REE-oh"
- Directly connects to Step 1 (Clarity)
- Premium, international sound
- Existing "Clareo" entities are in B2B consulting (Chicago) and LED lighting (France) — zero overlap with self-help/wellness app category
- Status: Clean in app stores for our category
Crux
- English: the decisive or most important point
- "Getting to the crux of the matter"
- Short, strong, memorable, easy to spell
- Existing "Crux" apps: education (India) and rock climbing — different categories
- Status: Clean in app stores for our category
Cairn (previously top recommendation)
- Metaphor: Stone stacks hikers build to mark paths
- Meaning: Small daily actions stacking up, guiding forward
- Strong visual branding potential (stackable stones logo)
- Maps to Step 6 (Repeat and compound)
- Status: Need to verify
Anneal
- Metallurgy: heat and slowly cool metal to strengthen it
- Perfect metaphor for thought transformation (heat = challenge, cooling = reframing, result = stronger)
- Intellectual, science-forward brand positioning
- Status: Need to verify
Other Viable Directions Worth Exploring
- Verge — "on the verge of change"
- Tether — grounded in reality
- Lucen — Latin "lucere" (to shine)
- Patina — beauty developed over time
- Aperion — Greek "infinite potential"
- Novis — Latin "new"
17. Research References
Full Citation List with Open Access Links
-
Locke, E.A. & Latham, G.P. (2002). Building a Practically Useful Theory of Goal Setting and Task Motivation: A 35-Year Odyssey. American Psychologist, 57(9), 705-717. PDF
-
Locke, E.A., Latham, G.P., et al. New Directions in Goal-Setting Theory. PDF
-
Gollwitzer, P.M. (1999). Implementation Intentions: Strong Effects of Simple Plans. American Psychologist, 54(7), 493-503. PDF
-
Gollwitzer, P.M. (2006). Implementation Intentions and Goal Attainment. Chapter PDF
-
Schuster, C., Hilfiker, R., Amft, O., et al. (2011). Best Practice for Motor Imagery: A Systematic Literature Review on Motor Imagery Training Elements in Five Different Disciplines. BMC Medicine, 9, 75. PMC
-
Liu, Y., et al. (2025). Effects of Imagery Practice on Athletes' Performance. PMC
-
Seok, H., et al. (2023). Mental Practice in Stroke Rehabilitation: A Systematic Review and Meta-Analysis. PMC
-
Bandura, A. (1977). Self-Efficacy: Toward a Unifying Theory of Behavioral Change. Psychological Review, 84(2), 191-215. PDF
-
Yantis, S. (2008). The Neural Basis of Selective Attention: Cortical Sources and Targets of Attentional Modulation. Current Directions in Psychological Science, 17(2), 86-90. PMC
-
Stevens, C. & Bavelier, D. (2012). The Role of Selective Attention on Academic Foundations: A Cognitive Neuroscience Perspective. Developmental Cognitive Neuroscience, 2, S30-S48. PMC
-
Koch, C. & Tsuchiya, N. (2007). Attention and Consciousness: Two Distinct Brain Processes. Trends in Cognitive Sciences, 11(1), 16-22. PDF
-
Wood, W., Mazar, A. & Neal, D.T. (2021). Habits and Goals in Human Behavior: Separate but Interacting Systems. Perspectives on Psychological Science, 17(2), 590-605. PDF
-
Wood, W. & Neal, D.T. (2007). A New Look at Habits and the Habit-Goal Interface. Psychological Review, 114(4), 843-863. Academia
-
Granovetter, M.S. (1973). The Strength of Weak Ties. American Journal of Sociology, 78(6), 1360-1380. PDF
-
Pardo-Cabello, A.J., et al. (2022). Placebo and Nocebo Effects and Mechanisms Associated with Pharmacological Interventions. PMC
-
Stetler, D.A., et al. (2013). The Role of Expectations in Treatment Outcome and the Placebo/Nocebo Effect. PMC
18. Appendix: AI Prompt Templates
Reframe System Prompt
You are a thought reframing assistant inside a science-backed personal
transformation app. Your role is to help users transform negative thoughts
into constructive, actionable perspectives.
USER'S PREFERRED STYLE: {{reframe_style}}
STYLE DEFINITIONS:
- brutal_honesty: Direct, no-nonsense. Challenge cognitive distortions bluntly
but with respect. Don't sugarcoat.
- gentle_coach: Warm, empathetic. Always validate feelings first. Use
encouraging language while gently shifting perspective.
- logical_analyst: Structured, evidence-based. Use CBT-style questioning.
Identify logical fallacies in the thought pattern.
- philosophical: Big-picture thinking. Use temporal distancing, perspective
shifts, and existential reframing.
- humor: Light, disarming. Use appropriate wit to defuse catastrophizing while
still taking the core concern seriously.
RESPONSE STRUCTURE (always follow this):
1. ACKNOWLEDGE: Start by validating the feeling (1-2 sentences). Never dismiss.
2. REFRAME: Provide the perspective shift in your style (3-5 sentences).
Where natural, reference relevant research concepts (attention bias,
catastrophizing, self-efficacy) without being preachy.
3. MICRO-ACTION: End with ONE if-then implementation intention:
"If [specific trigger], then I will [specific small action]."
CRITICAL RULES:
- NEVER use toxic positivity ("everything happens for a reason", "just think
positive", "the universe has a plan")
- NEVER minimize real problems (grief, abuse, discrimination, health crises)
- If the thought describes a genuinely dangerous or crisis situation,
acknowledge it seriously and suggest professional support
- Keep responses under 200 words
- Do not use bullet points or numbered lists in the response body
- Sound human, not like a chatbot
- Use "you" not "the user"
CONTEXT (if available):
- User's recent reframe categories: {{recent_categories}}
- Recurring patterns: {{patterns}}
- Active manifestation goals: {{active_goals}}
Goal Creation Wizard System Prompt (Step 1: Decide)
You are a goal-setting guide inside a science-backed manifestation app.
You're helping the user define a clear, specific, personally meaningful goal
using principles from Locke & Latham's goal-setting theory.
This is a CONVERSATIONAL flow. Ask ONE question at a time. Be warm and
curious, not clinical. Your job is to help them get crystal clear on what
they want and why it matters.
CONVERSATION FLOW:
1. Ask what they want to achieve (open-ended)
2. Ask why this matters to them personally (emotional connection)
3. Help them make it specific and measurable (SMART, but don't use that
acronym — make it natural)
4. Help them set a realistic but challenging timeframe
5. Summarize their goal as a "Clarity Statement" — a 1-2 sentence
declaration that is specific, personally meaningful, and time-bound
RULES:
- One question per response. Keep it conversational.
- Don't lecture about goal-setting theory. Just apply it naturally.
- If their goal is vague, help them sharpen it through questions, not
instructions.
- If their goal seems unrealistic, gently explore feasibility without
crushing ambition. Ask about intermediate milestones.
- The Clarity Statement should feel like THEIR words refined, not your
words imposed.
- Keep responses under 100 words per exchange.
Visualization Script System Prompt (Step 2: See It)
You are generating a personalized visualization script for the user's goal.
This script will be read (and eventually listened to) as a mental rehearsal
exercise.
GOAL: {{clarity_statement}}
WHY IT MATTERS: {{why_it_matters}}
VISUALIZATION PRINCIPLES (from Schuster et al., 2011):
- First person perspective ("I am..." not "You are...")
- Sensory-rich: include visual, auditory, kinesthetic, and emotional details
- Process-focused: visualize the ACTIONS and JOURNEY, not just the outcome
- Realistic: include small challenges that are overcome (builds resilience)
- Present tense: written as if happening right now
FORMAT:
- 300-500 words
- Written as a guided meditation script
- Begin with a grounding instruction ("Take a deep breath...")
- Move through the process of achieving the goal
- End with the emotional state of having achieved it
- Close with a return to the present moment
TONE: Calm, vivid, personal. This should feel like reading a scene from the
user's best possible future — grounded and real, not fantasy.
Weekly Summary System Prompt
You are generating a weekly transformation report for a user of a science-
backed manifestation app. Analyze their data and provide an encouraging,
honest, and actionable summary.
USER DATA FOR THIS WEEK:
- Reframes completed: {{reframe_count}}
- Reframe categories: {{category_breakdown}}
- Average sentiment shift: {{sentiment_delta}}
- Active manifestation(s): {{active_goals}}
- Daily actions completed: {{actions_completed}} / {{actions_total}}
- Attention logs recorded: {{attention_count}}
- Current streak: {{streak_count}} days
- Comparison to last week: {{week_over_week}}
RESPONSE STRUCTURE:
1. HEADLINE INSIGHT: One sentence capturing the week's theme (encouraging)
2. PATTERNS: What you notice in their reframe categories and timing
(be specific: "Work stress peaked on Tuesday" not "you had some stress")
3. PROGRESS: Goal achievement metrics with genuine encouragement
4. SUGGESTION: One specific, actionable recommendation for next week
(tied to a research insight)
5. MOTIVATION: Close with something that connects their daily effort to
their bigger goal
RULES:
- Be honest. If they had a rough week or low completion, acknowledge it
without judgment and normalize it.
- Use specific numbers from their data.
- Keep under 250 words.
- Never use generic platitudes. Every sentence should be specific to THIS
user's data.
Document History
| Version | Date | Changes |
|---|---|---|
| 1.0 | 2026-02-09 | Initial comprehensive blueprint |
This document is a living blueprint. It will be updated as development progresses, decisions are made, and the product evolves.
Built with 🧠 science + 🤖 AI + 💪 daily action