run-query · RawTree MCP Server MCP: Risk & Policy

WHAT IT DOES

What run-query does on RawTree MCP Server

AI agents invoke run-query to trigger actions in RawTree MCP 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.

WHY IT NEEDS A POLICY

Why run-query needs a policy

Although the tool claims to enforce read-only semantics and reject unsafe statements, it still executes arbitrary SQL queries provided by the caller. SQL injection, resource exhaustion attacks (expensive queries causing DoS), or exploitation of database-specific vulnerabilities could occur if the validation is bypassed or incomplete.

From the tool's definition Tool description states it 'Execute[s] a read-only SQL query against the configured RawTree database' and explicitly disclaims write operations ('NOT for: Inserting, updating, deleting, or mutating data').

Attacks that exploit this kind of access

Recommended rule Ready

Verdict

run-query Rate-limited

run-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.

View as policy code

policy.json

{
  "version": "1",
  "default": "deny",
  "tools": {
    "run-query": {
      "limits": [
        {
          "counter": "run-query_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 RawTree MCP Server, so you can run agents against it safely.

FAQ

Questions about run-query

What does the run-query tool do? +

Purpose: Execute a read-only SQL query against the configured RawTree database and return JSON rows, column metadata, statistics, and hints. NOT for: Inserting, updating, deleting, or mutating data. RawTree validates queries as read-only and rejects unsafe statements. Returns: RawTree. It is categorised as a Execute tool in the RawTree MCP Server 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-query? +

Register the RawTree MCP Server MCP server in PolicyLayer and add a rule for run-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 RawTree MCP Server. Nothing to install.

What risk level is run-query? +

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

Can I rate-limit run-query? +

Yes. Add a rate_limit block to the run-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.

How do I block run-query completely? +

Set action: deny in the PolicyLayer policy for run-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.

What MCP server provides run-query? +

run-query is provided by the RawTree MCP Server MCP server (rawtreedb/rawtree-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.