ChromaDB Remote MCP Server
This gives Claude (Desktop, Mobile, and Code) shared access to a self-hosted ChromaDB vector database over HTTP. You get all the standard CRUD operations for collections, embeddings, and semantic search wrapped as MCP tools, plus a REST API proxy for direct Python or JavaScript access. The architecture is straightforward: a Node.js gateway handles auth and protocol translation between MCP clients and your ChromaDB instance. Useful if you want persistent vector memory that syncs across all your Claude clients without round-tripping through Anthropic's servers. Ships as a Docker image with one-command setup, token auth, and Swagger docs at /docs. The main win is keeping your embeddings on your own infrastructure while maintaining the same semantic search context everywhere.
- GitHub stars
- 12
- tools
- 30
- auth
- Secret needed
- transport
- HTTP
- env vars
- 2
- last updated
- Oct 31, 2025
Tools
30 toolsPublic tool metadata for what this MCP can expose to an agent.
chroma_list_collectionsList all collections (with `limit` / `offset`)
List all collections (with `limit` / `offset`)
No parameter schema in public metadata yet.
chroma_create_collectionCreate a new collection (`configuration` / `schema` optional)
Create a new collection (`configuration` / `schema` optional)
No parameter schema in public metadata yet.
chroma_get_or_create_collectionIdempotent create-or-get (v2.2.0)
Idempotent create-or-get (v2.2.0)
No parameter schema in public metadata yet.
chroma_modify_collectionRename / change metadata or configuration (v2.2.0)
Rename / change metadata or configuration (v2.2.0)
No parameter schema in public metadata yet.
chroma_delete_collectionDelete a collection
Delete a collection
No parameter schema in public metadata yet.
chroma_get_collection_infoGet collection metadata
Get collection metadata
No parameter schema in public metadata yet.
chroma_get_collection_countGet document count (`read_level` optional)
Get document count (`read_level` optional)
No parameter schema in public metadata yet.
chroma_count_collectionsTotal collection count (v2.2.0)
Total collection count (v2.2.0)
No parameter schema in public metadata yet.
chroma_peek_collectionPreview collection contents
Preview collection contents
No parameter schema in public metadata yet.
chroma_add_documentsAdd documents (with `uris` for multi-modal)
Add documents (with `uris` for multi-modal)
No parameter schema in public metadata yet.
chroma_upsert_documentsIdempotent insert-or-update (v2.2.0)
Idempotent insert-or-update (v2.2.0)
No parameter schema in public metadata yet.
chroma_query_documentsSemantic search (with `query_uris` / `ids` pre-filter)
Semantic search (with `query_uris` / `ids` pre-filter)
No parameter schema in public metadata yet.
chroma_get_documentsRetrieve documents (`read_level` optional)
Retrieve documents (`read_level` optional)
No parameter schema in public metadata yet.
chroma_update_documentsUpdate existing documents (with `embeddings` / `uris`)
Update existing documents (with `embeddings` / `uris`)
No parameter schema in public metadata yet.
chroma_delete_documentsDelete by `ids` and/or `where` / `where_document` filter
Delete by `ids` and/or `where` / `where_document` filter
No parameter schema in public metadata yet.
chroma_heartbeatServer heartbeat (nanosecond timestamp)
Server heartbeat (nanosecond timestamp)
No parameter schema in public metadata yet.
chroma_get_server_versionServer version string
Server version string
No parameter schema in public metadata yet.
chroma_get_max_batch_sizeMax batch size (for client-side splitting)
Max batch size (for client-side splitting)
No parameter schema in public metadata yet.
chroma_get_user_identityCurrent tenant + databases
Current tenant + databases
No parameter schema in public metadata yet.
chroma_searchHybrid dense + sparse search (RRF).
Hybrid dense + sparse search (RRF).
No parameter schema in public metadata yet.
chroma_fork_collectionZero-copy fork (segment-level operation on object storage — architecturally requires the distributed compactor/storage stack).
Zero-copy fork (segment-level operation on object storage — architecturally requires the distributed compactor/storage stack).
No parameter schema in public metadata yet.
chroma_get_fork_countFork metadata lookup (depends on the distributed metadata store).
Fork metadata lookup (depends on the distributed metadata store).
No parameter schema in public metadata yet.
chroma_get_indexing_statusWAL offset + compactor index progress (requires the distributed WAL/compactor services).
WAL offset + compactor index progress (requires the distributed WAL/compactor services).
No parameter schema in public metadata yet.
chroma_admin_create_database
No parameter schema in public metadata yet.
chroma_admin_get_database
No parameter schema in public metadata yet.
chroma_admin_list_databases
No parameter schema in public metadata yet.
chroma_admin_create_tenant
No parameter schema in public metadata yet.
chroma_admin_get_tenant
No parameter schema in public metadata yet.
chroma_reset_databaseReset entire database (irreversible)
Reset entire database (irreversible)
No parameter schema in public metadata yet.
chroma_admin_delete_databaseDelete a database (requires both flags)
Delete a database (requires both flags)
No parameter schema in public metadata yet.
Sponsors
Agent CaffeineSponsoredClose your lid, come back to finished work
SnugSponsoredThe clipboard manager for the AI eraPromote your product hereReach 125k+ visitors monthly
WaxSponsoredBlock anything with supervision, without erasing your iPhonePromote your product hereReach 125k+ visitors monthly
knobsSponsoredSee your site on any phone, any day, in any languageConfiguration
CHROMADB_MCP_URL*URL of your deployed ChromaDB Remote MCP server
MCP_AUTH_TOKENsecretAuthentication token for the MCP server (if enabled)
Readme
ChromaDB Remote MCP Server
A Streamable HTTP MCP (Model Context Protocol) server that provides remote access to ChromaDB for AI assistants like Claude. Enables semantic search and vector database operations from mobile devices and remote locations.
Note: This project uses MCP Streamable HTTP (2025-03-26 spec). SSE transport is deprecated.
Cross-Platform AI Memory Server
Details
- Repository
- github.com/meloncafe/chromadb-remote-mcp
- Categories
- Databases, AI & LLM Tools
- Package
devsaurus/chromadb-remote-mcp:latest- Transport
- HTTP
- Auth
- Required
- Registry
- MCP registry, active
Related Databases MCP servers
View all
Microsoft SQL Serveralyiox/mcp-mssqlMCP server for SQL Server metadata, parameterized queries, plan analysis, and writes opt-in.9stars
db-mcp (SQLite MCP Server)neverinfamous/db-mcpSQLite MCP server with OAuth 2.1, HTTP/SSE, 122 tools, and smart tool filtering0stars
Knowledge Fabricinvarcore/knowledge-fabricMulti-tenant hybrid RAG & evidence verification MCP server enforcing PostgreSQL RLS.8stars
AppFlowy MCP Serverrahulk644/appflowy-mcp-serverAppFlowy MCP server: edit Markdown, manage typed database rows, search and import/export.8stars
Che Apple Mail Mcpkiki830621/che-apple-mail-mcpApple Mail MCP — 44 tools, SQLite millisecond search, .emlx parser, batch ops5stars
AlphaFold Sovereign MCPsmaniches/alphafold-sovereign-mcpMCP server for AlphaFold and 8 other biomedical data sources with a local SQLite knowledge graph5stars