Skill: Fast Unified Memory

# Skill: Fast Unified Memory

A high-performance unified memory system that integrates OpenClaw memory with semantic memory storage using Ollama's nomic-embed-text model for ultra-fast embeddings.

Overview

This skill provides a unified memory layer that combines:

Features

Requirements

Installation

# Install Ollama first
curl -fsSL https://ollama.ai/install.sh | sh

# Pull the embedding model
ollama pull nomic-embed-text

# Start Ollama
ollama serve

Usage

Commands

# Search both memory systems
node fast-unified-memory.js search "your query"

# Add a memory
node fast-unified-memory.js add "User prefers concise responses"

# List all memories
node fast-unified-memory.js list

# Show system stats
node fast-unified-memory.js stats

Architecture

┌─────────────────────────────────────────────┐
│           FAST UNIFIED MEMORY                │
│                                             │
│  ┌─────────────┐    ┌─────────────┐        │
│  │   OpenClaw  │    │   Semantic  │        │
│  │   Memory    │    │   Memory    │        │
│  │ (files)     │    │  (vectors) │        │
│  └─────────────┘    └─────────────┘        │
│           ↓                  ↓              │
│    [Keyword Match]   [Cosine Similarity]   │
│                                             │
│        Unified Results (ranked)             │
└─────────────────────────────────────────────┘

Performance

| Metric | Value |

|--------|-------|

| Embedding generation | ~40ms |

| Vector search | ~50ms |

| File search | ~40ms |

| **Total search** | **~130ms** |

Configuration

The skill uses these defaults:

Files

Troubleshooting

**Ollama not running:**

ollama serve

**Model not found:**

ollama pull nomic-embed-text

**Port conflict:**

The skill assumes Ollama is on port 11434. Update the `OLLAMA_URL` constant if using a different port.

License

MIT