execute_cql

What execute_cql does on Ncp

AI agents invoke execute_cql to trigger actions in Ncp. 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 execute_cql is rated High

This tool allows execution of arbitrary database queries. While CQL can be used for benign reads, the tool's design permits unrestricted query execution, making it capable of destructive operations (DROP, DELETE, TRUNCATE) or unauthorized data modification. Classified as Execute rather than Destructive because the actual harm depends on the specific query passed as an argument.

From the tool's definition Tool name 'execute_cql' and description 'Execute CQL query' indicate execution of arbitrary CQL (Cassandra Query Language) queries, which can perform read, write, or destructive operations depending on query content.

The rule that runs execute_cql safely

PolicyLayer is an MCP gateway: it sits between your AI agents and Ncp, and checks every tool call against a rule you set before the call runs. Nothing changes on the server itself. For execute_cql, this is the rule to start with:

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

Questions about execute_cql