claude_code_query · MCP Coding Agents MCP: Risk & Policy

WHAT IT DOES

What claude_code_query does on MCP Coding Agents

AI agents invoke claude_code_query to trigger actions in MCP Coding Agents. 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 claude_code_query needs a policy

This tool executes Claude Code queries, which can run arbitrary code or shell commands through the Claude Code agent. The server description explicitly mentions 'built-in tools for command execution' and integrates AI coding agents capable of writing and running code. An AI agent could misuse this to execute unintended or harmful code, giving it a high severity rating.

From the tool's definition Execute a Claude Code query with real-time message streaming

Attacks that exploit this kind of access

Recommended rule Ready

Verdict

claude_code_query Rate-limited

claude_code_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": {
    "claude_code_query": {
      "limits": [
        {
          "counter": "claude_code_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 MCP Coding Agents, so you can run agents against it safely.

FAQ

Questions about claude_code_query

What does the claude_code_query tool do? +

Execute a Claude Code query with real-time message streaming. It is categorised as a Execute tool in the MCP Coding Agents MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.

How do I enforce a policy on claude_code_query? +

Register the MCP Coding Agents MCP server in PolicyLayer and add a rule for claude_code_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 Coding Agents. Nothing to install.

What risk level is claude_code_query? +

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

Can I rate-limit claude_code_query? +

Yes. Add a rate_limit block to the claude_code_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 claude_code_query completely? +

Set action: deny in the PolicyLayer policy for claude_code_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 claude_code_query? +

claude_code_query is provided by the MCP Coding Agents MCP server (kadreio/mcp-coding-agents). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.