AI agents invoke execute_query to trigger actions in MCP SQL 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.
SQL query execution is a classic Execute category action—it runs code (SQL) whose effects depend on the arguments (the query text). While the tool itself is neutral, a malicious or misprompted query could read sensitive data, modify records, or (if the connection has DROP/DELETE permissions) perform destructive operations.
From the tool's definition Tool name is 'execute_query' and description states 'Execute SQL queries'. The sibling tools on this server include destructive capabilities like batch processing and schema operations, indicating this server manages production SQL databases.
Documented attack patterns abuse exactly the kind of access execute_query gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and MCP SQL Server, and nothing reaches the server without passing your rules. This is the rule we recommend for execute_query:
{
"version": "1",
"default": "deny",
"tools": {
"execute_query": {
"limits": [
{
"counter": "execute_query_rate",
"window": "minute",
"max": 10,
"scope": "grant"
}
]
}
}
}execute_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.
Free to start. No card required.
Execute SQL queries. It is categorised as a Execute tool in the MCP SQL Server MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the MCP SQL Server MCP server in PolicyLayer and add a rule for execute_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 MCP SQL Server. Nothing to install.
execute_query is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.
Yes. Add a rate_limit block to the execute_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.
Set action: deny in the PolicyLayer policy for execute_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.
execute_query is provided by the MCP SQL Server MCP server (ryudg/mcp-sql). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from MCP SQL Server, add the rest of your stack, and see everything your agents can call. Then put policy on all of it.
Free to start. No card required.
11 MCP SQL Server tools catalogued and risk-classified — across an index of 43,000+ MCP servers.