Does it require a specific database setup?

Yes, it typically utilizes Neo4j for graph-based storage and Qdrant for vector-based semantic search to manage memories.

What happens if the memory server is offline?

The skill includes graceful degradation logic, meaning the voice assistant will continue to function normally without memory features rather than crashing.

Is this skill specific to Claude Code?

While it is a general integration pattern, this skill provides the specific guidance and implementation patterns for Claude Code to manage and debug Kymera Voice memory systems.

What is Kymera Voice Memory used for?

It enables voice assistants to store and retrieve conversation history, allowing them to provide more personalized and context-aware responses based on past interactions.

How do I configure the connection to the memory server?

You can configure the server location via the WW_URL environment variable or by passing the ww_url parameter during initialization.

Kymera Voice Memory

Name: Kymera Voice Memory
Author: astoreyai

byastoreyai

•

Data Science & ML

Integrates long-term memory capabilities into the Kymera Voice assistant using World Weaver for context-aware interactions.

About

Kymera Voice Memory is a specialized skill designed to bridge the gap between real-time voice interactions and persistent data storage. By leveraging the World Weaver integration, Neo4j, and Qdrant, this skill enables voice assistants to recall past conversations, retrieve relevant context for current queries, and learn from every user interaction. It provides a robust framework for context enhancement and episode storage, featuring built-in graceful degradation to ensure the assistant remains functional even if the backend memory services are offline. This skill is essential for developers building stateful, intelligent voice interfaces that require a deep understanding of user history.

Key Features

1 GitHub stars
Async factory initialization for high-performance voice workflows
Automatic conversation episode storage for long-term AI learning
Context enhancement via semantic and graph-based memory retrieval
Graceful degradation for offline or unavailable memory services
Seamless integration with Neo4j and Qdrant database backends

Use Cases

Implementing a persistent 'knowledge brain' for custom AI voice projects
Developing voice assistants that remember user preferences across sessions
Debugging memory retrieval logic within World Weaver integrations

About

Key Features

1 GitHub stars
Async factory initialization for high-performance voice workflows
Automatic conversation episode storage for long-term AI learning
Context enhancement via semantic and graph-based memory retrieval
Graceful degradation for offline or unavailable memory services
Seamless integration with Neo4j and Qdrant database backends

Use Cases

Implementing a persistent 'knowledge brain' for custom AI voice projects
Developing voice assistants that remember user preferences across sessions
Debugging memory retrieval logic within World Weaver integrations