web_evaluate · K-Personal MCP: Risk & Policy

WHAT IT DOES

What web_evaluate does on K-Personal MCP

AI agents invoke web_evaluate to trigger actions in K-Personal MCP. 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 web_evaluate needs a policy

This tool runs arbitrary JavaScript in a browser page context, which constitutes code execution. An AI agent could misuse it to exfiltrate data, manipulate the DOM maliciously, interact with web APIs, or perform other operations depending on the arguments passed. This places it firmly in the Execute category with high severity due to the broad blast radius of arbitrary JS execution.

From the tool's definition 'JS 표현식 실행' — executes JavaScript expressions in the page context and returns the result

Attacks that exploit this kind of access

Recommended rule Ready

Verdict

web_evaluate Rate-limited

web_evaluate 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": {
    "web_evaluate": {
      "limits": [
        {
          "counter": "web_evaluate_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 K-Personal MCP, so you can run agents against it safely.

FAQ

Questions about web_evaluate

What does the web_evaluate tool do? +

페이지 컨텍스트에서 JS 표현식 실행 후 결과 반환. It is categorised as a Execute tool in the K-Personal MCP MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.

How do I enforce a policy on web_evaluate? +

Register the K-Personal MCP server in PolicyLayer and add a rule for web_evaluate: 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 K-Personal MCP. Nothing to install.

What risk level is web_evaluate? +

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

Can I rate-limit web_evaluate? +

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

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

web_evaluate is provided by the K-Personal MCP server (lee30934-byte/k-personal-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.