cortex-complete · Snowflake Cortex AI MCP Server MCP

WHAT IT DOES

What cortex-complete does on Snowflake Cortex AI MCP Server

AI agents invoke cortex-complete to trigger actions in Snowflake Cortex AI MCP Server. 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 cortex-complete needs a policy

This tool executes LLM inference via Snowflake's Cortex Complete API, triggering external model operations. While it doesn't modify data directly, it runs external computational operations whose effects depend on the arguments passed.

From the tool's definition Simple LLM chat completion API using Cortex Complete

Attacks that exploit this kind of access

Recommended rule Ready

Verdict

cortex-complete Rate-limited

cortex-complete 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": {
    "cortex-complete": {
      "limits": [
        {
          "counter": "cortex-complete_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 Cortex AI MCP Server, so you can run agents against it safely.

FAQ

Questions about cortex-complete

What does the cortex-complete tool do? +

Simple LLM chat completion API using Cortex Complete. It is categorised as a Execute tool in the Snowflake Cortex AI MCP Server MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.

How do I enforce a policy on cortex-complete? +

Register the Snowflake Cortex AI MCP Server MCP server in PolicyLayer and add a rule for cortex-complete: 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 Cortex AI MCP Server. Nothing to install.

What risk level is cortex-complete? +

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

Can I rate-limit cortex-complete? +

Yes. Add a rate_limit block to the cortex-complete 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 cortex-complete completely? +

Set action: deny in the PolicyLayer policy for cortex-complete. 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 cortex-complete? +

cortex-complete is provided by the Snowflake Cortex AI MCP Server MCP server (sfc-gh-tgordonjr/mcp-snow-tg). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.