High Risk →

web_search

Search the web via 14 engines: google, youtube, amazon, bing, duckduckgo, reddit, github, stackoverflow, wikipedia, twitter, linkedin, facebook, instagram, tiktok. Call snapshot after to read results.

How to control web_search ↓

AI agents invoke web_search to trigger actions in Camofox. 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

This tool triggers real browser automation actions against external web services (14 search engines) using a stealth anti-detection browser. While the primary intent is read-like (searching), it actually executes browser operations externally via a REST API, which constitutes an Execute action. The anti-detection/stealth nature increases risk as it is designed to bypass bot detection, raising the severity to medium.

From the tool's definition Search the web via 14 engines... triggers external operations against remote search services using an anti-detection browser with stealth fingerprinting

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

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

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

web_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 Camofox — 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 web_search tool do? +

Search the web via 14 engines: google, youtube, amazon, bing, duckduckgo, reddit, github, stackoverflow, wikipedia, twitter, linkedin, facebook, instagram, tiktok. Call snapshot after to read results. It is categorised as a Execute tool in the Camofox 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_search? +

Register the Camofox MCP server in PolicyLayer and add a rule for web_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 Camofox. Nothing to install.

What risk level is web_search? +

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

Can I rate-limit web_search? +

Yes. Add a rate_limit block to the web_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 web_search completely? +

Set action: deny in the PolicyLayer policy for web_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 web_search? +

web_search is provided by the Camofox MCP server (redf0x1/camofox-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.

Enforce policy on every Camofox tool call.

Deterministic rules across all 47 Camofox tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

Free to start. No card required.

47 Camofox 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.