Chroma MCP Server

Example Use Cases

• 1

  Store and retrieve documents with semantic search to surface relevant content based on meaning rather than keyword overlap.

• 2

  Filter search results by structured metadata (e.g., year, field) while performing similarity searches.

• 3

  A persistent, local vector store with CRUD operations for managing documents and supporting offline or self-hosted workflows.

Description

This server implements the Model Context Protocol (MCP) using Chroma as a vector database to store documents with content and metadata, enabling semantic similarity search and filtered retrieval. It supports create/read/update/delete operations, metadata and content filtering, persistent local storage, and retry/error handling for robustness. The implementation is designed for integration with agents and tools that need fast semantic search and document management with on-disk persistence.