chromadb-memory

Overview

ChromaDB Memory is a long-term semantic memory skill that uses ChromaDB and local Ollama embeddings to store and retrieve memories. It provides auto-recall and manual search capabilities, allowing users to inject relevant context into their conversations.

Key Features

• Auto-recall: injects relevant context every turn
• chromadb_search tool: manual semantic search over your ChromaDB collection
• 100% local: Ollama (nomic-embed-text) for embeddings, ChromaDB for vector storage

How It Works

1. You send a message
2. Plugin embeds your message via Ollama (nomic-embed-text, 768 dimensions)
3. Queries ChromaDB for nearest neighbors
4. Results above minScore are injected into the agent's context as <chromadb-memories>
5. Agent responds with relevant long-term context available

Use Cases

• Conversational memory: store and retrieve memories from conversations
• Knowledge base: use ChromaDB as a knowledge base and Ollama for embeddings
• Personal assistant: use ChromaDB Memory to provide personalized recommendations and context