think · MCP Claude Code MCP: Risk & Policy

WHAT IT DOES

What think does on MCP Claude Code

AI agents call think as a supporting operation in MCP Claude Code workflows.

WHY IT NEEDS A POLICY

Why think needs a policy

With no description available, the tool's behavior cannot be determined. The name 'think' suggests an internal reasoning/scratchpad step with no side effects, which would classify as Other. However, confidence is very low due to the empty description. Given the server context (code analysis, file modification, command execution), it is unlikely to be destructive, but certainty is minimal.

From the tool's definition Tool name is 'think' with an empty description. No functional details are provided.

Attacks that exploit this kind of access

Recommended rule Ready

Verdict

think Rate-limited

think gets a rate cap, and everything else on the server is denied unless you say otherwise.

View as policy code

policy.json

{
  "version": "1",
  "default": "deny",
  "tools": {
    "think": {
      "limits": [
        {
          "counter": "think_rate",
          "window": "minute",
          "max": 60,
          "scope": "grant"
        }
      ]
    }
  }
}

SET A RULE FOR THIS TOOL →

Instant setup, no code required.

PolicyLayer is an MCP gateway: every tool call is checked against your rule before it reaches MCP Claude Code, so you can run agents against it safely.

FAQ

Questions about think

What does the think tool do? +

think. It is categorised as a Other tool in the MCP Claude Code MCP Server, which means it performs auxiliary operations.

How do I enforce a policy on think? +

Register the MCP Claude Code MCP server in PolicyLayer and add a rule for think: 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 Claude Code. Nothing to install.

What risk level is think? +

think is a Other tool with low risk. Read-only tools are generally safe to allow by default.

Can I rate-limit think? +

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

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

think is provided by the MCP Claude Code MCP server (sdglbl/mcp-claude-code). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.