Frequently Asked Questions about Chroma MCP Server
Q: What is Chroma MCP Server?
A: Chroma MCP Server is an implementation of the Model Context Protocol (MCP) that provides database capabilities for Chroma, an open-source embedding database. It allows AI models to access and interact with external data sources and tools.
Q: What is the Model Context Protocol (MCP)?
A: MCP is an open protocol designed for seamless integration between LLM applications and external data sources or tools, offering a standardized framework to provide LLMs with the context they require.
Q: What are the key features of Chroma MCP Server?
A: Key features include flexible client types (ephemeral, persistent, HTTP, cloud), collection management, document operations, support for various tools (chroma_list_collections, etc.), and embedding function support (e.g., cohere, openai).
Q: What client types does Chroma MCP Server support?
A: It supports Ephemeral (in-memory), Persistent (file-based), HTTP (for self-hosted Chroma instances), and Cloud clients (for Chroma Cloud integration).
Q: How do I add documents to a collection in Chroma MCP Server?
A: You can use the chroma_add_documents tool to add documents with optional metadata and custom IDs.
Q: How can I query documents using Chroma MCP Server?
A: You can use the chroma_query_documents tool to query documents using semantic search with advanced filtering.
Q: What embedding functions are supported by Chroma MCP Server?
A: Chroma MCP supports several embedding functions: default, cohere, openai, jina, voyageai, and roboflow.
Q: How do I configure Chroma MCP Server to use an API key for an embedding function?
A: Set the environment variable CHROMA_<EMBEDDING_FUNCTION>_API_KEY="<key>". For example, for Cohere, set CHROMA_COHERE_API_KEY="<your_cohere_api_key>".
Q: How do I integrate Chroma MCP Server with Claude Desktop?
A: Add a Chroma section to your claude_desktop_config.json file with the appropriate command and arguments for your client type (e.g., ephemeral, persistent, cloud, or HTTP).
Q: Can I use environment variables to configure the Chroma MCP Server client?
A: Yes, you can use environment variables. The server automatically loads variables from a .env file specified by --dotenv-path or from system environment variables.
Q: Where can I find demos of how to use Chroma MCP Server?
A: Reference usages, such as shared knowledge bases and adding memory to context windows, can be found in the Chroma MCP Docs.
Q: What is UBOS and how does it relate to Chroma MCP Server?
A: UBOS is a full-stack AI Agent Development Platform that simplifies the process of building, deploying, and managing AI agents. Hosting your Chroma MCP Server on UBOS provides advantages such as simplified orchestration, seamless integration, and enterprise-grade security.
Chroma MCP Server
Project Details
- tjwells47/chroma-mcp
- Apache License 2.0
- Last Updated: 4/9/2025
Recomended MCP Servers
This is an implementation project of a JVM-based MCP (Model Context Protocol) server. The project aims to provide...
Stock market data provider for Claude Desktop using MCP
New Linkedin MCP
Model Context Protocol (MCP) server that enables CRUD operations on a WebDAV endpoint with basic authentication.
A mcp server that bridges Dune Analytics data to AI agents.





