Analytics

Snowflake MCP Server

Connect your account, then chat with AI to run tools.

Explore Snowflake warehouses, databases, schemas, tables, and run SQL analytics workloads from chat.

Best forDataAnalytics EngineeringBI OperationsPlatform Administration

Try this workflow

Explore warehouse structure

List Snowflake databases, schemas, tables, and warehouses, then summarize the likely analytics domains available in this account.

Start chatting
Managed
28 tools
Agent guide included

Opens MCPBundles Studio with this server selected. After sign-in, chat and run tools from the same thread.

Browse all tools

What you can do with Snowflake

Built for

Data Teams, Analytics Engineering, BI Operations, Platform Administration

Example workflows

Explore warehouse structure

Uses discovery tools before writing SQL.

Try this

List Snowflake databases, schemas, tables, and warehouses, then summarize the likely analytics domains available in this account.

Describe a table before querying

Reduces SQL mistakes by inspecting schema first.

Try this

Describe this Snowflake table, explain the important columns and types, then draft a safe SELECT query for the business question.

Run an analytics query

Connects warehouse execution to business analysis.

Try this

Execute a Snowflake SQL query to summarize monthly revenue by customer segment, then explain the result and caveats.

Review warehouse usage context

Highlights compute boundaries before query execution.

Try this

List warehouses and explain which one should be used for an exploratory query versus a heavier analytics workload.

Context to know first

What Snowflake objects can agents inspect?

Agents can inspect warehouses, databases, schemas, tables, table columns, and execute SQL statements against the connected Snowflake account.

Why should agents list metadata before querying?

Snowflake accounts can contain many databases and schemas. Listing metadata first helps select the right warehouse, database, schema, and table before running SQL.

Can agents run arbitrary SQL?

The server executes SQL under the connected role and warehouse context. Prompts should be explicit about read-only analysis versus DDL or DML changes.

Related editorial

Snowflake Workflows with AI

How AI agents can run SQL, discover schemas, manage warehouses, and read query history to answer data questions and tune performance.

Read article

AI Skill
SKILL.md

Domain knowledge for Snowflake — workflow patterns, data models, and gotchas for your AI agent.

Snowflake

Cloud data warehouse with separation of storage and compute, addressed by your account URL. All operations go through the Snowflake SQL REST API — every tool ultimately executes SQL statements.

Data Model

  • Account — top-level container identified by an account URL (e.g., org-account.snowflakecomputing.com).
  • Warehouses — virtual compute clusters that execute queries. Can be started, suspended, and resized independently.
  • Databases — logical groupings of schemas. Each account can have many databases.
  • Schemas — namespaces within a database containing tables, views, stages, and other objects.
  • Tables — structured data storage. Snowflake supports standard, transient, temporary, and external tables.
  • Views — saved queries that act as virtual tables.
  • Stages — locations for data loading/unloading (internal or external like S3/GCS/Azure).

Key Workflows

  1. Start by listing databases and warehouses to understand what's available in the account.
  2. Explore schemas within a database to find tables of interest.
  3. Describe tables to understand column types and structure before querying.
  4. Execute SQL for any operation — SELECT, INSERT, CREATE, DDL, DML, or analytics queries.

Gotchas

  • Identifiers are uppercase by default: Unquoted identifiers in Snowflake are stored as uppercase. Use double quotes for case-sensitive names.
  • Warehouse must be running: Queries require an active warehouse. If suspended, it auto-resumes but adds latency.
  • Semi-structured data: Use VARIANT, OBJECT, ARRAY types with : and [] path notation for JSON-like data.
  • Time travel: SELECT ... AT(TIMESTAMP => ...) or BEFORE(STATEMENT => ...) for historical queries.

Tools in this Server (28)

Snowflake Account Info

Get current Snowflake session information: account name, user, role, warehouse, database, schema, and Snowflake version.

Snowflake Credit Usage

Get warehouse credit consumption over time. Shows credits used per warehouse per day — essential for cost monitoring and optimization.

Snowflake Describe Table

Describe the columns, types, and constraints of a Snowflake table.

Snowflake Execute

Execute SQL statements on Snowflake. Supports all SQL operations including SELECT, INSERT, UPDATE, DELETE, CREATE, DROP, ALTER, SHOW, DESCRIBE, and mo...

Snowflake Explain Query

Get the execution plan for a SQL query using EXPLAIN. Shows the operations Snowflake will perform, estimated costs, and data flow without actually run...

Snowflake Find Duplicates

Find duplicate rows in a Snowflake table based on specified columns. Returns the duplicate values and their occurrence counts.

Snowflake Get Table Ddl

Get the full CREATE TABLE statement (DDL) for a Snowflake table, including column definitions, clustering keys, and constraints.

Snowflake Get Table Stats

Get storage and clustering statistics for a Snowflake table: row count, size in bytes, active bytes, clustering depth and key, and time travel retenti...

Snowflake Get View Ddl

Get the SQL CREATE statement (DDL) for a Snowflake view, showing the underlying query and column definitions.

Snowflake List Databases

List all databases in the Snowflake account that the current role can access.

Snowflake List Functions

List user-defined functions (UDFs) in a Snowflake schema, including language and return type.

Snowflake List Procedures

List stored procedures in a Snowflake schema, including their arguments and language.

Snowflake List Roles

List all roles in the Snowflake account, showing ownership and granted-to counts.

Snowflake List Schemas

List all schemas in a Snowflake database.

Snowflake List Stages

List data stages (internal and external) in a Snowflake schema for data loading and unloading.

Snowflake List Tables

List all tables in a Snowflake schema with row counts and size information.

Snowflake List Views

List all views in a Snowflake schema including materialized views, with their definition status and owner.

Snowflake List Warehouses

List all virtual warehouses in the Snowflake account with their size and status.

Snowflake Login History

Get recent login history for the Snowflake account. Shows successful and failed login attempts with client types and IPs.

Snowflake Preview Data

Preview rows from a Snowflake table with optional column selection, filtering, and ordering. More flexible than sample — use for targeted data inspect...

Snowflake Profile Column

Profile a single column in a Snowflake table: row count, distinct values, nulls, min/max, and most common values with frequencies.

Snowflake Query History

Retrieve recent query history from the Snowflake account, including execution times and status.

Snowflake Resize Warehouse

Resize a Snowflake warehouse to change compute capacity. Larger warehouses process queries faster but consume more credits per second.

Snowflake Resume Warehouse

Resume a suspended Snowflake warehouse so it can execute queries. Warehouses auto-resume on query if configured, but explicit resume avoids the cold-s...

Snowflake Sample Data

Get a random sample of rows from a Snowflake table using TABLESAMPLE. Useful for understanding data shape and values without scanning full table.

Snowflake Storage Usage

Get storage usage history across the Snowflake account — average bytes stored, failsafe bytes, and stage bytes over time.

Snowflake Suspend Warehouse

Suspend a running Snowflake warehouse to stop credit consumption. Running queries will complete before suspension.

Snowflake Warehouse Status

Get detailed status and configuration of a Snowflake warehouse: state, size, auto-suspend/resume settings, queued and running queries.

Frequently Asked Questions

What is the Snowflake MCP server?

Snowflake MCP server for cloud data warehouse operations. Execute SQL queries, explore databases and schemas, inspect table structures, manage warehouses, and run analytics workloads against your Snowflake account. It provides 28 tools that AI agents can use through the Model Context Protocol (MCP).

How do I connect Snowflake to my AI agent?

Add the MCPBundles server URL to your MCP client configuration (Claude Desktop, Cursor, VS Code, etc.). The URL format is: https://mcp.mcpbundles.com/bundle/snowflake. Authentication is handled automatically.

How many tools does Snowflake provide?

Snowflake provides 28 tools that can be called by AI agents, along with a SKILL.md that gives your AI agent domain knowledge about when and how to use them.

What authentication does Snowflake require?

Snowflake uses API Key. Snowflake requires credentials. Connect via MCPBundles and authentication is handled automatically.

What Snowflake objects can agents inspect?

Agents can inspect warehouses, databases, schemas, tables, table columns, and execute SQL statements against the connected Snowflake account.

Why should agents list metadata before querying?

Snowflake accounts can contain many databases and schemas. Listing metadata first helps select the right warehouse, database, schema, and table before running SQL.

Can agents run arbitrary SQL?

The server executes SQL under the connected role and warehouse context. Prompts should be explicit about read-only analysis versus DDL or DML changes.

What Snowflake gotchas matter?

Identifiers are uppercased by default, warehouses must be available for queries, and semi-structured data requires Snowflake's VARIANT path syntax.

Setup Instructions

Connect Snowflake to any MCP client in minutes

MCP URLhttps://mcp.mcpbundles.com/bundle/snowflake
Claude.aiCursorVS CodeVS Code InsidersGooseLM StudioCherry StudioChatGPT

One-click install:

Install on Claude.aiInstall for organization (admins)

The link prefills the Add custom connector dialog — you still review the values and click Add, then Connect to complete OAuth.

Or add manually

  1. Open claude.ai → Settings → Connectors.
  2. Click the + button and choose Add custom connector.
  3. Set Name to Snowflake and paste the MCP URL into Remote MCP server URL.
  4. Click Add. Snowflake will appear under Not connected — select it and click Connect to complete OAuth.
Name: Snowflake
Remote MCP server URL: https://mcp.mcpbundles.com/bundle/snowflake
Authentication: OAuth

Custom connectors at claude.ai require a paid Claude plan (Pro, Max, Team, or Enterprise).

Related MCP Servers

More analytics integrations you might like

1Forge

1Forge provides real-time and historical financial market data including forex exchange rates, crypt...

2 tools
Analytics
OSS Insight

OSSInsight provides deep analytics and insights for open source projects on GitHub, powered by TiDB ...

18 tools
Analytics
Open Data DC

Open Data DC provides access to a variety of datasets related to the District of Columbia's municipa...

3 tools
Analytics
FullStory

Fullstory is a digital experience analytics platform that captures user interaction data on websites...

24 tools
Analytics
Grafana

Grafana MCP server for observability and monitoring. Search dashboards, explore datasources, query a...

39 tools
Analytics
Zoho Analytics

Zoho Analytics metadata API provides access to metadata information for reports and dashboards in Zo...

9 tools
Analytics

Ready to use Snowflake?

Sign in to connect your credentials and start running tools from the chat.

Snowflake MCP Server & Skill — 28 Tools