🧠 OMNI-MEMORY v2.0

The Ultimate Memory System for OpenCode AI

Neural Memory Weaving™ | Quantum Search™ | Multi-Modal Fusion | Knowledge Graph

🌟 What Makes OMNI-MEMORY Special

OMNI-MEMORY ist nicht nur ein Memory-Plugin – es ist ein kognitives Erweiterungssystem, das dein Wissen wie ein zweites Gehirn organisiert.

🔮 Einzigartige Features

| Feature | Beschreibung | |---------|-------------| | Neural Memory Weaving™ | Memories verknüpfen sich automatisch wie Synapsen im Gehirn | | Quantum Search™ | 5-in-1 Hybrid-Suche: Semantic + Keyword + Graph + Temporal + Fuzzy | | Memory Decay/Reinforcement | Vergessenskurve wie beim Menschen – wichtige Memories stärken sich | | Knowledge Graph | Visuelle Darstellung deiner Wissensnetzwerke | | Autonomous Curator | Hintergrund-Agent der automatisch aufräumt, merged, taggt | | Digital Twin | Lernt deinen Coding-Stil für personalisierte AI-Antworten | | Memory Marketplace | Kaufe/Verkaufe Wissenspakete (React Patterns, Security, etc.) | | Multi-Modal Fusion | Text + Code + Screenshots (OCR) + Voice Notes | | E2E Cloud Sync | Verschlüsselte Synchronisation zwischen Geräten | | Temporal Layers | "Wie sah mein Wissen vor 1 Monat aus?" |

📊 Vergleich

| Feature | FREE | PRO ($29) | TEAM ($199) | |---------|------|-----------|-------------| | Memories | 500 | ∞ | ∞ | | Keyword Search | ✅ | ✅ | ✅ | | Semantic Search | ❌ | ✅ | ✅ | | Knowledge Graph | ❌ | ✅ | ✅ | | Auto-Curator | ❌ | ✅ | ✅ | | Weekly Digest | ❌ | ✅ | ✅ | | Cloud Sync | ❌ | ✅ | ✅ | | Multi-Modal | ❌ | ✅ | ✅ | | Memory Marketplace | ❌ | ✅ | ✅ | | Team Sharing | ❌ | ❌ | ✅ | | SSO/ Audit Logs | ❌ | ❌ | ✅ | | Priority Support | ❌ | ✅ | ✅ |

🚀 Installation

opencode plugin install @xanomyrox/omni-memory

Manuelle Installation

git clone https://github.com/xanomyrox/opencode-memory.git
cd opencode-memory
npm install
npm run build

📋 Commands

Core Commands

| Command | Shortcut | Beschreibung | |---------|----------|-------------| | /omni-extract | Ctrl+E | Extrahiert Memories aus Conversation | | /omni-memory | Ctrl+M | Browse alle Memories | | /omni-search | Ctrl+K | Quantum Search (5-in-1) | | /omni-graph | – | Knowledge Graph anzeigen | | /omni-health | – | Memory Health Check | | /omni-digest | – | Weekly Digest | | /omni-filter | – | Nach Typ filtern | | /omni-delete | – | Memory löschen | | /omni-export | Ctrl+X | Export (JSON, MD, DOT) | | /omni-stats | – | Statistiken |

Advanced Commands

| Command | Beschreibung | |---------|-------------| | /omni-semantic | Semantic Search (AI-powered) | | /omni-voice | Voice Note aufnehmen | | /omni-ocr | Text aus Screenshot extrahieren | | /omni-activate | License aktivieren | | /omni-license | License Status | | /omni-upgrade | Upgrade zu PRO | | /omni-marketplace | Memory Marketplace | | /omni-help | Hilfe |

🔑 License Keys

Format

OMNI-PRO-XXXX-XXXX-XXXX-XXXX
OMNI-TEAM-XXXX-XXXX-XXXX-XXXX

Aktivierung

/omni-activate <license-key>

Purchase

• PRO: https://xanomyrox.gumroad.com/omni-memory-pro
• TEAM: https://xanomyrox.gumroad.com/omni-memory-team

🧠 Neural Memory Weaving™

OMNI-MEMORY erstellt automatisch Verbindungen zwischen verwandten Memories:

// Auto-generated links basierend auf:
// - Semantic similarity (Vektor-Ähnlichkeit)
// - Shared entities (gleiche Funktionen, Classes, etc.)
// - Temporal proximity (kurz nacheinander erstellt)
// - User-defined relationships

{
  id: "mem_abc123",
  title: "React useEffect Dependency Array",
  links: [
    { target: "mem_def456", type: "semantic", score: 0.87 },
    { target: "mem_ghi789", type: "shares_entity", entities: ["useEffect"] },
    { target: "mem_jkl012", type: "temporal", score: 0.65 }
  ]
}

🔍 Quantum Search™

Die 5-in-1 Hybrid-Suche kombiniert:

1. Semantic (35%): Vektor-basierte Bedeutungssuche
2. Keyword (25%): Exakte Wortübereinstimmung
3. Graph (20%): Verbundene Memories
4. Temporal (10%): Zeitbezogene Suche
5. Fuzzy (10%): Unscharfe Übereinstimmung

Natural Language Queries

"similar patterns to useEffect optimization"
"memories connected to authentication"
"last week's bugfixes"
"exactly 10 results for testing"

📈 Memory Health

OMNI-MEMORY berechnet einen Health Score basierend auf:

• Quantity (20%): Wie viele Memories?
• Quality (30%): Importance & Strength
• Connectivity (20%): Wie gut vernetzt?
• Diversity (15%): Typen-Vielfalt
• Freshness (15%): Regelmäßige Updates

/omni-health
# Output: Score 72/100
# Recommendations:
# - Review 15 weak memories
# - Add more testing documentation
# - Link 23 isolated memories

🗄️ Storage

Local Storage

• Memories: ~/.config/opencode/data/omni-memory.json
• Vectors: ~/.config/opencode/data/omni-vectors.json
• License: ~/.config/opencode/data/omni-license.json
• Digests: ~/.config/opencode/data/digests/

Cloud Storage (PRO+)

• E2E verschlüsselt mit AES-GCM (256-bit)
• Supabase Backend
• CRDT-basierte Konfliktlösung

🎯 Memory Types

| Type | Icon | Beschreibung | |------|------|-------------| | bugfix | 🐛 | Bug fixes, workarounds, solutions | | feature | ✨ | New features, requirements | | concept | 💡 | Abstract concepts, ideas | | code_pattern | 📝 | Coding conventions, patterns | | preference | ⭐ | User preferences, style | | fact | 📌 | Facts, notes, information | | testing | 🧪 | Test strategies, patterns | | config | ⚙️ | Configuration, setup | | project | 📁 | Project-specific info | | security | 🔒 | Security patterns, auth | | documentation | 📄 | Docs, guides | | deployment | 🚀 | CI/CD, Docker, K8s |

🤖 Autonomous Curator

Der Background-Agent läuft alle 15 Minuten:

Aufgaben

1. Merge Duplicates: Erkennt und fusioniert Duplikate (>95% similarity)
2. Auto-Tag: Taggt ungetaggte Memories basierend auf Keywords
3. Clean Obsolete: Löscht alte, schwache, unwichtige Memories
4. Auto-Link: Erstellt neue Verbindungen zwischen related Memories

Weekly Digest

Jede Woche generiert der Agent einen Report:

Period: 2026-03-26 - 2026-04-02
Total Memories: 47

By Type:
  bugfix: 15
  feature: 12
  concept: 10
  testing: 10

Top Memories:
  • React useEffect Optimization (strength: 94%)
  • PostgreSQL Index Patterns (strength: 89%)

Insights:
  • You focused heavily on bugfix this week (15 memories)
  • Great knowledge weaving! 67% of memories are interconnected

Recommendations:
  • Add more testing documentation
  • Review 5 weak memories

🔐 Security

Encryption

• Algorithm: AES-GCM (256-bit)
• Key Derivation: PBKDF2 (100.000 Iterationen)
• Hash: SHA-256

License Validation

• Checksum-Validierung (Modulo 97)
• Device-Bindung
• Optional: Online-Verifizierung

🏪 Memory Marketplace

Featured Packs

| Pack | Price | Memories | Rating | |------|-------|----------|--------| | React Best Practices | $9.99 | 50 | ⭐ 4.8 | | Node.js Architecture | $14.99 | 75 | ⭐ 4.9 | | Security Checklist | $12.99 | 40 | ⭐ 4.7 | | Testing Strategies | $11.99 | 60 | ⭐ 4.6 | | DevOps Essentials | $15.99 | 80 | ⭐ 4.8 |

Submit Your Own Pack

// Pack Struktur
{
  id: "pack_react_patterns",
  name: "React Best Practices",
  description: "Essential React patterns",
  price: 9.99,
  memories: [...],
  author: "your_username"
}

🛠️ Development

Build

npm install
npm run build

Test

npm test

Lint

npm run lint

📦 API

MemoryStore

// Create
const result = store.create({
  title: "My Memory",
  content: "Content here",
  type: "concept",
  importance: 7,
  tags: ["react", "hooks"]
});

// List
const memories = store.list({
  type: "bugfix",
  limit: 50,
  sort: "importance"
});

// Search
const results = await search.search("useEffect optimization", {
  useSemantic: true,
  useGraph: true,
  limit: 20
});

// Update
store.update(id, { importance: 9 });

// Delete
store.delete(id);

// Reinforce
store.reinforce(id, 0.1); // +10% strength

KnowledgeGraph

// Get stats
const stats = graph.getStats();
// { nodes: 150, edges: 340, clusters: 12, density: "15.2%" }

// Get neighbors
const neighbors = graph.getNeighbors(memoryId, 2); // depth=2

// Find paths
const paths = graph.getPaths(sourceId, targetId, 5);

// Export
const dot = graph.exportGraph('dot');

Analytics

// Health check
const health = await analytics.calculateHealth();
// { score: 72, factors: {...}, recommendations: [...] }

// Growth tracking
const growth = analytics.getKnowledgeGrowth(30);

// Type distribution
const distribution = analytics.getTypeDistribution();

🎨 Customization

Settings

{
  "autoExtract": true,
  "autoSync": true,
  "maxExtractionsPerSession": 50,
  "confidenceThreshold": 0.5,
  "decayRate": 1.0,
  "maxContextMemories": 5,
  "minRelevanceScore": 0.6
}

🐛 Troubleshooting

###常见问题

Q: Embedding engine lädt nicht

Lösung: @xenova/transformers ist optional.
Feature-extraction funktioniert auch ohne.

Q: Cloud Sync verbindet nicht

Lösung: PRO/TEAM License erforderlich.
Supabase URL und Key konfigurieren.

Q: Knowledge Graph zeigt nichts

Lösung: vis-network installieren:
npm install vis-network

Q: License wird nicht erkannt

Lösung: Format prüfen: OMNI-PRO-XXXX-XXXX-XXXX-XXXX
/omni-license für Status

📝 Changelog

v2.0.0 (2026-04-02)

Major Release – Next-Gen Features

• ✨ Neural Memory Weaving™ (auto-linking)
• 🧠 Quantum Search™ (5-in-1 hybrid)
• 🕸️ Knowledge Graph with visualization
• 🤖 Autonomous Curator Agent
• 📰 Weekly Digest
• ☁️ E2E Cloud Sync (Supabase)
• 📸 Multi-Modal (OCR, Voice)
• 🏪 Memory Marketplace
• 💪 Memory Decay/Reinforcement
• 📊 Analytics & Health Score
• 🔐 Enhanced license system

v1.0.1 (2026-01-15)

• Basic memory storage
• Keyword search
• Manual extraction
• JSON export

📄 License

MIT License – Copyright (c) 2026 xanomyrox

See LICENSE for full text.

🙏 Support

• Issues: https://github.com/xanomyrox/opencode-memory/issues
• Discord: https://discord.gg/omni-memory
• Twitter: @xanomyrox
• Email: [email protected]

🌟 Credits

Built with ❤️ by xanomyrox

Powered by:

• @xenova/transformers
• vis-network
• Tesseract.js
• Supabase

Made for developers who think bigger. 🚀