read_query · Snowflake MCP Server (Read-Only) MCP

WHAT IT DOES

What read_query does on Snowflake MCP Server (Read-Only)

AI agents invoke read_query to trigger actions in Snowflake MCP Server (Read-Only). What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call — builds kicked off, notifications sent, workflows started.

WHY IT NEEDS A POLICY

Why read_query needs a policy

Although described as read-only and limited to SELECT, the tool executes SQL against a live database. Poorly crafted or injected queries (e.g., expensive queries, queries exposing sensitive data at scale) could have significant blast radius. The act of executing SQL is categorized as Execute rather than Read, since the effect depends on the query arguments passed.

From the tool's definition 'Execute a SELECT query' — the tool runs SQL queries against a Snowflake database

Attacks that exploit this kind of access

Recommended rule Ready

Verdict

read_query Rate-limited

read_query stays usable, but rate-capped — a runaway agent can't fire it dozens of times a minute. Everything else on the server is denied unless you say otherwise.

View as policy code

policy.json

{
  "version": "1",
  "default": "deny",
  "tools": {
    "read_query": {
      "limits": [
        {
          "counter": "read_query_rate",
          "window": "minute",
          "max": 10,
          "scope": "grant"
        }
      ]
    }
  }
}

RATE-LIMIT THIS TOOL →

Instant setup, no code required.

PolicyLayer is an MCP gateway: every tool call is checked against your rule before it reaches Snowflake MCP Server (Read-Only), so you can run agents against it safely.

FAQ

Questions about read_query

What does the read_query tool do? +

Execute a SELECT query. It is categorised as a Execute tool in the Snowflake MCP Server (Read-Only) MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.

How do I enforce a policy on read_query? +

Register the Snowflake MCP Server (Read-Only) MCP server in PolicyLayer and add a rule for read_query: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Snowflake MCP Server (Read-Only). Nothing to install.

What risk level is read_query? +

read_query is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.

Can I rate-limit read_query? +

Yes. Add a rate_limit block to the read_query rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.

How do I block read_query completely? +

Set action: deny in the PolicyLayer policy for read_query. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.

What MCP server provides read_query? +

read_query is provided by the Snowflake MCP Server (Read-Only) MCP server (sgriffin-magnoliacap/mcp-snowflake-server-readonly). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.