arcticdb-mcp
Allows using MinIO as a storage backend for ArcticDB, enabling structured read/write access to time-series data stored in S3-compatible buckets via the ArcticDB MCP server.
This server does not currently have a deployable release on Glama. Follow the repository's setup instructions to install and configure it for your MCP client, including any required credentials and local dependencies.
Overview
arcticdb-mcp turns ArcticDB into an AI data copilot. Assistants can automate symbol updates, batch jobs, and time-series queries while built-in versioning protects every change. Go from question to data change in minutes, not scripts, with full traceability. You can also design your own workflows on top of arcticdb-mcp.
Related MCP server: Sablier MCP Server
Table of Contents
Demo Video
Watch the live watch db demo:
https://github.com/user-attachments/assets/dd81f229-e1f9-46e6-8e79-7c7252bc25c6
Quickstart
Install and run with
uvx(recommended):
uvx arcticdb-mcpConfigure your MCP client (Claude Desktop / Cursor / Windsurf / Continue):
{
"mcpServers": {
"arcticdb": {
"command": "uvx",
"args": ["arcticdb-mcp"],
"env": {
"ARCTICDB_URI": "lmdb:///path/to/your/database"
}
}
}
}Ask your assistant:
"Show me the last 5 rows of AAPL in library finance"
"Create a snapshot of finance before I update symbols"
"List versions for symbol ES_intraday"
Installation
Choose one:
# Recommended
uvx arcticdb-mcp
# Or
pipx install arcticdb-mcp
# Or
pip install arcticdb-mcpPython requirement: >=3.9
Configuration
Set ARCTICDB_URI in MCP client config or environment.
URI examples
Local LMDB (Linux/macOS):
lmdb:///path/to/dbLocal LMDB (Windows):
lmdb://C:/path/to/dbAWS S3:
s3://s3.amazonaws.com:bucket?region=us-east-1&access=KEY&secret=SECRETAzure Blob:
azure://AccountName=X;AccountKey=Y;Container=ZS3-compatible (MinIO, etc.):
s3://your-endpoint:bucket?access=KEY&secret=SECRET
You can also use a .env file:
ARCTICDB_URI=lmdb:///path/to/dbRun Modes
stdio (default)
This is the default mode used by desktop MCP clients.
ARCTICDB_URI=lmdb:///path/to/db python -m arcticdb_mcpHTTP / SSE
Set ARCTICDB_MCP_PORT to run over HTTP/SSE:
Use an integer from 1 to 65535. An unset or empty value selects stdio.
ARCTICDB_URI=lmdb:///path/to/db ARCTICDB_MCP_PORT=8000 python -m arcticdb_mcpEndpoint:
http://localhost:8000/sseTools
Current server exposes 49 tools.
Arctic and Libraries
Tool | File | What it does |
|
| Return the ArcticDB URI used by this MCP server connection. |
|
| Return a compact summary of the ArcticDB store. |
|
| Modify a configurable library option. |
|
| List all libraries in the Arctic instance. |
|
| Create a new library with the given name. |
|
| Delete a library and all its underlying data permanently. |
|
| Check whether a library with the given name exists. |
|
| Return the name and full list of symbols stored in a library. |
|
| Return non-enterprise library options. |
|
| Return enterprise library options. |
Symbols
Tool | File | What it does |
|
| List all symbol names stored in the given library. |
|
| Check whether a symbol exists in the given library. |
|
| Read a symbol's full data as row records with index included. |
|
| Read paginated slice of a symbol with backend row-slicing, column projection, and version consistency. |
|
| Return the first |
|
| Return the last |
|
| Write data to a symbol, creating a new version. |
|
| Append rows to an existing symbol, creating a new version. |
|
| Update a date range of a timeseries symbol, creating a new version. |
|
| Delete a symbol and all its versions from the library. |
|
| Delete rows within a date range from a timeseries symbol, creating a new version. |
|
| Return schema, row count, date range, and metadata summary for a symbol. |
|
| List all versions of a symbol with version number and write timestamp. |
|
| Read only metadata for a symbol without loading its data. |
|
| Update metadata for a symbol without modifying its data. |
Batch Operations
Tool | File | What it does |
|
| Write multiple symbols in one call. |
|
| Append to multiple symbols in one call. |
|
| Update multiple symbols in one call. |
|
| Delete multiple symbols or symbol versions in one call. |
|
| Read multiple symbols in one call. |
|
| Read metadata for multiple symbols in one call. |
|
| Get schema/description for multiple symbols in one call. |
|
| Write metadata for multiple symbols in one call. |
|
| Write arbitrary pickle-serializable data to a symbol. |
|
| Write multiple pickle-serializable payloads in one call. |
Snapshots
Tool | File | What it does |
|
| Create a named snapshot of the current library state. |
|
| List the names of all snapshots in a library. |
|
| Delete a named snapshot from a library. |
|
| Read a symbol as it existed when a snapshot was created. |
Query Helpers
Tool | File | What it does |
|
| Filter rows using conditions. |
|
| Filter rows where a column value is in a list. |
|
| Group by a column and aggregate. |
|
| Filter rows by date range on a datetime-indexed symbol. |
|
| Resample a datetime-indexed symbol and aggregate. |
Maintenance
Tool | File | What it does |
|
| Force a symbol-list cache reload for a library. |
|
| Compact symbol-list cache into a single key in storage. |
|
| Check whether compaction would reduce segment count for a symbol. |
|
| Defragment symbol data by compacting fragmented row-sliced segments. |
|
| Remove non-snapshotted symbol versions except the latest one. |
Paginated Symbol Reads
Use read_symbol_page to inspect large symbols a page at a time without loading the full dataset into memory or overflowing assistant response context:
Parameters:
library,symbol,offset(default 0),limit(default 100, max 1000), optionalcolumns, optionalas_of.Backend Slicing: Row ranges and column projections are pushed directly to ArcticDB storage engines.
Version Pinning: The initial page resolves the symbol's version (
version). Passas_of=versionon subsequent pages to guarantee consistent reads across mutations.Continuation Info: Each response includes
count,total_rows,has_more, andnext_offset.
Two-Page Example
# Page 1: Read the first 100 rows of symbol AAPL
page1 = read_symbol_page(library="finance", symbol="AAPL", offset=0, limit=100)
# Returns: {"version": 2, "offset": 0, "limit": 100, "count": 100, "total_rows": 250, "has_more": True, "next_offset": 100, "data": [...]}
# Page 2: Pin version=2 and use next_offset to fetch the next slice
page2 = read_symbol_page(
library="finance",
symbol="AAPL",
offset=page1["next_offset"],
limit=100,
as_of=page1["version"],
)
# Returns: {"version": 2, "offset": 100, "limit": 100, "count": 100, "total_rows": 250, "has_more": True, "next_offset": 200, "data": [...]}Example Prompts
"Read symbol NVDA from library finance"
"Filter NVDA where price > 500 and volume >= 1000"
"Resample ES_intraday to 5min and aggregate price mean, volume sum"
"Write metadata owner=research to symbol ES_intraday"
"Delete data in range for ES_intraday from 2024-01-01 to 2024-01-05"
Development
git clone https://github.com/YMuskrat/arcticdb-mcp
cd arcticdb-mcp
pip install -e .Run locally:
ARCTICDB_URI=lmdb:///path/to/db python -m arcticdb_mcpTest with MCP Inspector:
ARCTICDB_URI=lmdb:///tmp/test_db npx @modelcontextprotocol/inspector python -m arcticdb_mcpCommunity Backlog
These ArcticDB capabilities are intentionally left open for contributors:
stagefinalize_staged_datasort_and_finalize_staged_datadelete_staged_dataget_staged_symbolsread_batch_and_joinadmin_tools
Contributing
See CONTRIBUTING.md.
License
MIT
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
This server cannot be deployed
Maintenance
Related MCP Connectors
Trading MCP server for AI agents, with live market data, account reads and controlled execution.
Hosted MCP server for AI-driven data ops. Create apps, manage schemas, and CRUD structured data.
Research-only MCP server: your AI as a quant research desk. 90 tools, no trades, no brokers.
MCP server that lets your AI assistant build and deploy apps, hosting each one at a live URL.
Related MCP Servers
- AlicenseBqualityDmaintenanceMCP server for Lakehouse42, enabling code-first tool discovery, hybrid search, document management, and Iceberg time-travel queries with optimized responses.107 npmMIT

Sablier MCP Serverofficial
AlicenseAqualityDmaintenanceAn MCP server that lets AI assistants analyze portfolios, stress-test scenarios, generate synthetic market paths, and scan SEC filings — in under 2 minutes.833MIT- AlicenseAqualityDmaintenanceMCP server for CyberQuant data sharing platform that enables AI assistants to directly query and analyze financial data using CSV output for token efficiency.7MIT
- AlicenseAqualityDmaintenanceAn MCP server that gives AI assistants direct, structured access to China A-share market data.3Apache 2.0