Chroma is a Databases MCP server that lets Claude Code, Cursor, Windsurf and any MCP-compatible AI agent embeddings, vector search, and document storage with Chroma AI database for RAG applications. Install in 1 minute with mcpizy install chroma.
Databases
Embeddings, vector search, and document storage with Chroma AI database for RAG applications.
Official homepagemcpizy install chromauvx chroma-mcplist_collectionsList Chroma collections
create_collectionCreate a new collection
Inputs
namestringrequiredadd_documentsAdd documents with optional embeddings
Inputs
collectionstringrequireddocumentsstring[]requiredidsstring[]requiredquery_collectionSemantic search in a collection
Inputs
collectionstringrequiredquery_textsstring[]requiredn_resultsnumberoptionaldelete_documentsDelete documents by ID
Inputs
collectionstringrequiredidsstring[]requiredWorks identically across clients. Only the config file path differs.
~/.claude.json{
"mcpServers": {
"chroma": {
"command": "uvx",
"args": [
"chroma-mcp"
],
"env": {
"CHROMA_CLIENT_TYPE": "persistent",
"CHROMA_DATA_DIR": "/path/to/chroma-data"
}
}
}
}.cursor/mcp.json{
"mcpServers": {
"chroma": {
"command": "uvx",
"args": [
"chroma-mcp"
],
"env": {
"CHROMA_CLIENT_TYPE": "persistent",
"CHROMA_DATA_DIR": "/path/to/chroma-data"
}
}
}
}~/.codeium/windsurf/mcp_config.json{
"mcpServers": {
"chroma": {
"command": "uvx",
"args": [
"chroma-mcp"
],
"env": {
"CHROMA_CLIENT_TYPE": "persistent",
"CHROMA_DATA_DIR": "/path/to/chroma-data"
}
}
}
}Chroma runs locally — no credentials needed for the persistent client
Paste any of these prompts into Claude Code, Cursor or another MCP-compatible client.
“Create a 'knowledge-base' collection and index these 100 documents”
Uses: create_collection, add_documents
“Semantic search my knowledge base for 'refund policy'”
Uses: query_collection
“List every collection and its document count”
Uses: list_collections
If Chroma doesn't fit your stack, these Databases MCP servers solve similar problems.
The Chroma MCP server is an Databases Model Context Protocol server that lets Claude Code, Cursor, Windsurf, VS Code with Copilot, and other MCP-compatible AI agents embeddings, vector search, and document storage with Chroma AI database for RAG applications. It exposes Chroma's capabilities as tools the AI can call directly from your editor or CLI.
The fastest way is the MCPizy CLI: run `mcpizy install chroma` and MCPizy will add the server to your `.claude.json` automatically. You can also install it manually by adding an entry under `mcpServers` in `.claude.json` with the command `uvx chroma-mcp` and restarting Claude Code.
Yes. The Chroma MCP server is free and open source (see the GitHub repository linked on this page). You may still need a Chroma account or API key to connect the server to the underlying service, but the MCP layer itself has no MCPizy subscription cost.
Yes. Any MCP-compatible client works — including Claude Code, Claude Desktop, Cursor (via `.cursor/mcp.json`), Windsurf, VS Code with Copilot Chat, and custom agents built on the MCP SDK. The same install command targets all of them; only the config file path differs.
Once installed, your AI agent can embeddings, vector search, and document storage with Chroma AI database for RAG applications directly inside your conversation. Typical use cases include asking Claude Code or Cursor to run Chroma operations, inspect results, chain Chroma with other MCP servers (see our Workflow Recipes), and automate repetitive databases tasks without leaving your editor.