WakeIQX Playground

Experience the 5-layer temporal intelligence architecture with interactive scenarios. Each demo shows how WakeIQX tracks context across time dimensions: Past (WHY), Present (HOW), Future (WHAT), Adaptive (HOW WELL), and Personality (HOW FRAMED).

🎮 WakeIQX Playground

Experience 4-layer temporal intelligence with interactive scenarios

🔍 Layer 1: Past (WHY)💾 Layer 2: Present (HOW)🔮 Layer 3: Future (WHAT)🔄 Layer 4: Adaptive (HOW WELL)

🔍

Development Session - Causal Chain

See how Layer 1 (Causality Engine) tracks decision history backwards through time

💾

Memory Tier Evolution

Watch how Layer 2 (Memory Manager) manages context lifecycle over time

🔮

Future Context Prediction

See Layer 3 (Propagation Engine) predict which contexts you'll need next

⚡

Save Context & Reconstruct Reasoning

Create a new context and later understand WHY it was created

🔄

Adaptive Weights — Layer 4 in Action

See how Layer 4 (Meta-Learning) tunes prediction weights based on your project's actual access patterns

🔎

Semantic Search Across Time

Find contexts by keyword and see how they connect across all 4 layers

About These Examples

All scenarios above are based on real MCP tool implementations with:

Actual tool schemas from the WakeIQX codebase
Real 5-layer architecture (Causality, Memory, Propagation, Meta-Learning, Personality Modes)
Authentic responses matching actual tool output format

What You're Seeing:

User Query - Natural language question about context
Tool Invocation - MCP server selects appropriate tool and parameters
Tool Result - Temporal intelligence analysis across layers
WakeIQX Response - Actionable insights with causal chains, memory tiers, and predictions

The 5-Layer Brain Architecture:

┌─────────────────────────────────────────────────────┐
│  Layer 5: PERSONALITY MODES (Presentation)          │
│  historian · prophet · archaeologist · minimalist   │
└─────────────────────────────────────────────────────┘
                         ↑
┌─────────────────────────────────────────────────────┐
│  Layer 4: META-LEARNING (Adaptive - HOW WELL)       │
│  Tunes prediction weights per project over time     │
└─────────────────────────────────────────────────────┘
                         ↑
┌─────────────────────────────────────────────────────┐
│  Layer 3: PROPAGATION ENGINE (Future - WHAT)        │
│  Predicts which contexts will be needed next        │
└─────────────────────────────────────────────────────┘
                         ↑
┌─────────────────────────────────────────────────────┐
│  Layer 2: MEMORY MANAGER (Present - HOW)            │
│  Manages context relevance and lifecycle            │
└─────────────────────────────────────────────────────┘
                         ↑
┌─────────────────────────────────────────────────────┐
│  Layer 1: CAUSALITY ENGINE (Past - WHY)             │
│  Tracks decision history and reasoning chains       │
└─────────────────────────────────────────────────────┘

Key Features Demonstrated:

🔍 Layer 1: Causality Engine - Backward causal chain reconstruction
💾 Layer 2: Memory Manager - Memory tier evolution (ACTIVE → RECENT → ARCHIVED → EXPIRED)
🔮 Layer 3: Propagation Engine - Multi-factor prediction scoring (temporal + causal + frequency)
🔄 Layer 4: Meta-Learning - Per-project adaptive weight tuning based on real outcomes
🎭 Layer 5: Personality Modes - historian, prophet, archaeologist, minimalist
⚡ Cross-Layer Integration - How all five layers work together
📊 Temporal Analytics - Statistics and insights across time dimensions

Memory Tier System:

Tier	Age	Purpose	Example
🔥 ACTIVE	< 1 hour	Hot, frequently accessed	Current work session
⚡ RECENT	1-24 hours	Warm, recently used	Today's context
📦 ARCHIVED	1-30 days	Cold, aging	Last week's work
❄️ EXPIRED	> 30 days	Pruning candidates	Old contexts

Prediction Scoring Algorithm:

Combined Score = (Temporal × w₁) + (Causal × w₂) + (Frequency × w₃)

Default weights: w₁ = 0.40, w₂ = 0.30, w₃ = 0.30

Temporal: Recent access patterns and momentum
Causal: Position in causal chain (root causes score higher)
Frequency: Overall popularity and access count

Weights adapt per project via Layer 4 Meta-Learning after 20+ recorded outcomes. Each project learns its own optimal blend — see the Adaptive Weights scenario above.

Want to Try the Real Thing?

To use WakeIQX with your own projects:

Install Claude Desktop - Required for MCP integration
Clone the Repo - Open source on GitHub
Follow Setup Guide - 5-minute configuration with Cloudflare D1

Real-World Use Cases

🔍 Development Session Tracking

typescript

// Save context with causality
save_context({
  project: "my-app",
  summary: "Implemented OAuth2 authentication",
  context: "Chose OAuth2 over JWT for third-party integrations",
  causedBy: "ctx_security_discussion_123",
  actionType: "implementation",
  rationale: "Need social login providers"
})

// Later: Why did I choose OAuth2?
reconstruct_reasoning({
  snapshotId: "ctx_oauth_implementation_456"
})
// → See the full causal chain back to original discussion

💾 Memory Management

typescript

// Check memory health
get_memory_stats({
  project: "my-app"
})
// → See tier distribution, identify expired contexts

// Clean up old contexts
prune_expired_contexts({
  limit: 50
})
// → Remove contexts older than 30 days

🔮 Proactive Context Loading

typescript

// Get prediction scores
get_high_value_contexts({
  project: "my-app",
  minScore: 0.6,
  limit: 10
})
// → Pre-fetch likely-needed contexts before starting work

// Update predictions
update_predictions({
  project: "my-app",
  staleThreshold: 3600 // 1 hour
})
// → Recalculate scores for fresh recommendations

⚡ Full Workflow Example

typescript

// Morning: Start work session
const recentContexts = await load_context({
  project: "my-app",
  limit: 5
})

// Identify what you were working on
const causalChain = await build_causal_chain({
  snapshotId: recentContexts[0].id
})

// Get predictions for today's work
const predictions = await get_high_value_contexts({
  project: "my-app",
  minScore: 0.7
})

// Work throughout the day, saving contexts...

// Evening: Check memory health
const stats = await get_memory_stats({
  project: "my-app"
})

Technical Deep Dive

Interested in how the 5-layer architecture works under the hood?

Architecture Overview - Complete system design
Layer 1: Causality Engine - Decision history tracking
Layer 2: Memory Manager - Lifecycle management
Layer 3: Propagation Engine - Prediction scoring
Layer 4: Meta-Learning - Adaptive weight tuning
Layer 5: Personality Modes - Temporal postures
Tools Documentation - All 15 MCP tools
API Reference - Service interfaces and entities

Ready to add temporal intelligence to your AI workflows?

Past → Present → Future → Adaptive → Personality | Temporal Intelligence for AI Agents 🐦

Get Started →

WakeIQX Playground ​