AI agents call mysql_query to retrieve information from Mysql without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool is restricted to SELECT queries and related read operations that do not modify, delete, or execute arbitrary code. The safety validation and allowlist of statement types prevent Write, Execute, and Destructive actions. The blast radius from misuse is minimal since queries can only retrieve existing data, not alter system state.
From the tool's definition Tool description explicitly states 'Run a read-only MySQL query' with 'Allowed statements: SELECT, WITH SELECT, SHOW, DESCRIBE/DESC, and safe EXPLAIN'. These are all read-only query types that retrieve or inspect data without modification.
Attacks that exploit this kind of access
Run a read-only MySQL query. Allowed statements: SELECT, WITH SELECT, SHOW, DESCRIBE/DESC, and safe EXPLAIN. It is categorised as a Read tool in the Mysql MCP Server, which means it retrieves data without modifying state.
Register the Mysql MCP server in PolicyLayer and add a rule for mysql_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 Mysql. Nothing to install.
mysql_query is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the mysql_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 mysql_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.
mysql_query is provided by the Mysql MCP server (xujcbiz/mysql-mcp-server). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →