Run a raw DML query, like INSERT, UPDATE, DELETE, etc.
AI agents call run_dml_query to permanently remove resources in Pgsql — typically in cleanup and lifecycle workflows. It does its job in a single call, and there is no undo.
DML queries include DELETE operations which permanently remove data, as well as INSERT and UPDATE. Since DELETE is irreversible and the tool accepts raw DML (giving an agent full ability to DELETE or mass-UPDATE arbitrary data), the most severe applicable category is Destructive. The blast radius is critical as a misused query could wipe entire tables.
From the tool's definition Run a raw DML query, like INSERT, UPDATE, DELETE, etc.
Documented attack patterns abuse exactly the kind of access run_dml_query gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and Pgsql, and nothing reaches the server without passing your rules. This is the rule we recommend for run_dml_query:
{
"version": "1",
"default": "deny",
"hide": [
"run_dml_query"
]
}run_dml_query disappears from the agent's tool list entirely, and any attempt to call it is denied. The rest of the server keeps working.
Free to start. No card required.
Run a raw DML query, like INSERT, UPDATE, DELETE, etc. It is categorised as a Destructive tool in the Pgsql MCP Server, which means it can permanently delete or destroy data. Block by default and require explicit approval.
Register the Pgsql MCP server in PolicyLayer and add a rule for run_dml_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 Pgsql. Nothing to install.
run_dml_query is a Destructive tool with critical risk. Critical-risk tools should be blocked by default and only enabled with explicit human approval.
Yes. Add a rate_limit block to the run_dml_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 run_dml_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.
run_dml_query is provided by the Pgsql MCP server (twn39/pgsql-mcp-server). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Start from Pgsql, 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.
9 Pgsql tools catalogued and risk-classified — across an index of 43,000+ MCP servers.