How is this different from connecting Snowflake MCP directly?

Connected directly, an agent sees every tool the server exposes and holds the credential itself. Through TrustGate you choose which tools each agent gets, and every call is logged against the agent that made it.

What if the Snowflake MCP API changes?

We track the Snowflake MCP schema, so the tool catalogue stays current. Tools you have already allowed keep working, and anything new stays denied until you enable it.

Can I give different agents different Snowflake tools?

Yes. Access is set per agent: one can get a handful of read tools, another the full set. You grant the whole MCP or single tools within it.

Where can I see what my agents did in Snowflake?

Every call is logged against the agent that made it, with the tool it used. Activity can be sent on to your own monitoring.

Do I need to run the Snowflake MCP server myself?

No. TrustGate connects to the Snowflake MCP server and sits in front of it. Your agent only talks to your TrustGate URL, never to Snowflake directly.

Connect Snowflake MCP to your AI agents, securely.

Q: Is Snowflake MCP secure?

The server itself exposes 16 tools, and some of them can write to your Snowflake account. Behind TrustGate an agent only sees the tools you allow, requests and responses are inspected, and every call is logged. Anything outside the allowlist is denied before it reaches Snowflake.

One secured endpoint for every Snowflake call your AI agents make, scoped per user, inspected, and logged.

Get started for free

Get a demo

Try Snowflake MCP now

What your agents can ask once Snowflake MCP is connected.

“Search the support documents for this error code”
“What were last quarter's revenues by region?”
“Describe the schema of this table”

“List the semantic views available to me”
“Run this query against the sales database”
“Which tables exist in this schema?”

Select the tools your AI agents can call

16 tools from the Snowflake MCP schema.

Showing 16 tools

cortex_searchreadQuery unstructured data in Snowflake, as commonly used in Retrieval Augmented Generation applications.
cortex_analystreadQuery structured data in Snowflake via rich semantic modeling.
cortex_agentreadAgentic orchestrator across structured and unstructured data retrieval.
list_semantic_viewsreadList the semantic views available in a database or schema.
describe_semantic_viewreadDescribe the structure of a semantic view, including its tables, dimensions and metrics.
query_semantic_viewreadQuery a semantic view with dimensions, metrics and filters.
show_semantic_dimensionsreadShow all semantic dimensions in the account, database, or schema.
show_semantic_metricsreadShow all semantic metrics in the account, database, or schema.
get_semantic_view_ddlreadGet the DDL for a semantic view.
write_semantic_view_query_toolreadWrites a query statement for a semantic view using dimensions, metrics or facts, without running it.
create_objectwriteCreate a Snowflake object such as a database, schema, table, view, warehouse, stage, role or user.
create_or_alter_objectwriteCreate a Snowflake object, or alter it if it already exists.
describe_objectreadDescribe a Snowflake object and its properties.
list_objectsreadList Snowflake objects of a given type within a database or schema.
drop_objecthigh riskDrop a Snowflake object. The object and its contents are removed.
run_snowflake_querywriteRun LLM-generated SQL, managed by the statement permissions configured for the server.

What Snowflake MCP exposes

Snowflake MCP groups its 16 tools into 4 categories, each with a suggested starting policy.

Category	Covers	Access	Suggested policy
Cortex AI	Cortex Search over unstructured data, Cortex Analyst over semantic models, and Cortex Agent orchestration	Read	Scope each configured service
Semantic Views	Semantic view discovery, description and querying	Read	Limit access to intended models
Object Management	Create, create or alter, describe, list and drop common Snowflake objects	Read / Write	Require approval for drops; keep object manager disabled unless needed
SQL	Execution of generated SQL under configured statement permissions	Read / Write	Allow only the statement types you have explicitly permitted