Datris MCP Server

MCP (Model Context Protocol) server for the Datris AI Agent-Native Data Platform. Enables AI agents (Claude Desktop, Claude Code, Cursor, and custom frameworks) to natively interact with the platform — discover data, create pipelines, upload files, monitor jobs, search vector databases, query structured data, and answer questions with AI.

Install

pip install datris-mcp-server

Usage

stdio mode (Claude Desktop / Claude Code)

DATRIS_API_URL=http://localhost:8080 datris-mcp-server

Or run directly:

DATRIS_API_URL=http://localhost:8080 python server.py

SSE mode (Docker / remote agents)

DATRIS_API_URL=http://localhost:8080 datris-mcp-server --sse --port 3000

Docker

The MCP server starts automatically with docker compose up in SSE mode on port 3000.

Configuration

Claude Desktop

Add to ~/Library/Application Support/Claude/claude_desktop_config.json:

{
  "mcpServers": {
    "datris": {
      "command": "datris-mcp-server",
      "env": {
        "DATRIS_API_URL": "http://localhost:8080"
      }
    }
  }
}

Claude Code

Add to .mcp.json in your project root:

{
  "mcpServers": {
    "datris": {
      "command": "datris-mcp-server",
      "env": {
        "DATRIS_API_URL": "http://localhost:8080"
      }
    }
  }
}

Environment Variables

Variable	Default	Description
`DATRIS_API_URL`	`http://localhost:8080`	Datris REST API server URL
`REQUIRE_API_KEY`	`false`	Reject SSE/HTTP sessions that connect without `x-api-key`

Authentication

There is no server-side API key. Each connecting agent authenticates per session by sending an x-api-key header; mcp-server forwards that key as-is to the Datris REST API on every tool call. The Datris REST API validates against oss/api-keys Vault secret (single-tenant) or api-key-mappings (multi-tenant). Manage keys in the Configuration UI's Secrets tab.

Tools

30+ tools across these categories:

Pipeline Management — create, list, get, delete pipelines; upload files; monitor jobs
Vector Search — semantic search across Qdrant, Weaviate, Milvus, Chroma, pgvector
Database Query — read-only SQL queries (PostgreSQL) and MongoDB queries
Metadata Discovery — explore databases, schemas, tables, columns, collections
AI — RAG-powered question answering
System — health checks, version info

See the full documentation at docs.datris.ai/mcp-server.

License

Apache 2.0

Release files for datris-mcp-server 1.20.1

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Source distribution (sdist)

Source distribution for datris-mcp-server 1.20.1
File	Size	Uploaded
datris_mcp_server-1.20.1.tar.gz	68.3 kB	Aug 25, 2026	Details

Built distribution (wheel)

Table of built distributions (wheels) for datris-mcp-server 1.20.1
File	Interpreter	ABI	Platform	Reset
datris_mcp_server-1.20.1-py3-none-any.whl 69.0 kB Aug 25, 2026	Python 3	none	any	Details

Total release size: 137.3 kB

Release files / datris_mcp_server-1.20.1.tar.gz

Download URL	datris_mcp_server-1.20.1.tar.gz
Size	68.3 kB
Tags	Source
SHA-256 checksum How to use checksums	`456e9751f70a4866224424322a8f41c5277933a6a65fde989533f5c8968cba12`
BLAKE2b-256 checksum How to use checksums	`34a71c087eedbb6db2560d708b47fec95baf454774bbd33869342b9b90faaa1d`
Upload date	Aug 25, 2026
Uploaded using Trusted Publishing? What is trusted publishing?	No
Uploaded via	`twine/7.0.0 CPython/3.14.6`

Release files / datris_mcp_server-1.20.1-py3-none-any.whl

Download URL	datris_mcp_server-1.20.1-py3-none-any.whl
Size	69.0 kB
Tags	Python 3
SHA-256 checksum How to use checksums	`edcafece34bdd9579b5cf6f1b1253cb5a6d154b02aa9f906e03cc90b94ad58d0`
BLAKE2b-256 checksum How to use checksums	`36cde698a670716a756538aaceeddb32cc09d47be7831c7a95796dfa3a4b865a`
Upload date	Aug 25, 2026
Uploaded using Trusted Publishing? What is trusted publishing?	No
Uploaded via	`twine/7.0.0 CPython/3.14.6`