High Risk →

search

Perform a Google search by navigating to https://www.google.com/search?q=encodedQuery using the provided query text.

How to control search ↓

AI agents invoke search to trigger actions in Steel 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.

High Risk

Although the intent is a read-like search, the tool executes a browser action (navigation) via Puppeteer, which constitutes an external operation. It fits Execute rather than Read because it drives a real browser session with side effects (network requests, session state changes). Severity is medium since web navigation can expose session data or be chained with other tools for more harmful actions.

From the tool's definition navigating to https://www.google.com/search?q=encodedQuery — this triggers an external browser navigation action via Puppeteer

Documented attack patterns abuse exactly the kind of access search gives an agent:

PolicyLayer is an MCP gateway — it sits between your AI agents and Steel MCP Server, and nothing reaches the server without passing your rules. This is the rule we recommend for search:

policy.json
{
  "version": "1",
  "default": "deny",
  "tools": {
    "search": {
      "limits": [
        {
          "counter": "search_rate",
          "window": "minute",
          "max": 10,
          "scope": "grant"
        }
      ]
    }
  }
}

search 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.

  1. Create a free account and register Steel MCP Server — nothing to install.
  2. Add this policy — paste it, or build it visually.
  3. Point your MCP client (Claude, Cursor, anything) at your gateway URL.
RATE-LIMIT THIS TOOL →

Free to start. No card required.

Go deeper

What does the search tool do? +

Perform a Google search by navigating to https://www.google.com/search?q=encodedQuery using the provided query text. It is categorised as a Execute tool in the Steel 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 search? +

Register the Steel MCP Server MCP server in PolicyLayer and add a rule for search: 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 Steel MCP Server. Nothing to install.

What risk level is search? +

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

Can I rate-limit search? +

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

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

search is provided by the Steel MCP Server MCP server (steel-dev/steel-mcp-server). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.

Enforce policy on every Steel MCP Server tool call.

Deterministic rules across all 9 Steel MCP Server tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

Free to start. No card required.

9 Steel MCP Server tools catalogued and risk-classified — across an index of 42,500+ MCP servers.

// GET IN TOUCH

Have a question or want to learn more? Send us a message.

Message sent.

We'll get back to you soon.