run_sql · Pg Mnemosyne MCP: Risk & Policy

WHAT IT DOES

What run_sql does on Pg Mnemosyne

AI agents invoke run_sql to trigger actions in Pg Mnemosyne. 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 run_sql needs a policy

Although this tool *could* be used destructively (DELETE/DROP) or for reads (SELECT), its primary risk is that it executes code (SQL) whose effects are entirely argument-dependent. The 'arbitrary' qualifier means an agent could invoke destructive or malicious queries. Execution of untrusted SQL against a live database is a critical risk.

From the tool's definition Tool description states it 'Executes arbitrary SQL queries and returns results as JSON.' The word 'arbitrary' indicates the tool accepts any SQL statement without restriction.

Attacks that exploit this kind of access

Recommended rule Ready

Verdict

run_sql Rate-limited

run_sql 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": {
    "run_sql": {
      "limits": [
        {
          "counter": "run_sql_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 Pg Mnemosyne, so you can run agents against it safely.

FAQ

Questions about run_sql

What does the run_sql tool do? +

Executes arbitrary SQL queries and returns results as JSON. It is categorised as a Execute tool in the Pg Mnemosyne MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.

How do I enforce a policy on run_sql? +

Register the Pg Mnemosyne MCP server in PolicyLayer and add a rule for run_sql: 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 Pg Mnemosyne. Nothing to install.

What risk level is run_sql? +

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

Can I rate-limit run_sql? +

Yes. Add a rate_limit block to the run_sql 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 run_sql completely? +

Set action: deny in the PolicyLayer policy for run_sql. 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 run_sql? +

run_sql is provided by the Pg Mnemosyne MCP server (janadasroor/pg-mnemosyne-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.