crawl_start · UnWeb MCP Server MCP: Risk & Policy

WHAT IT DOES

What crawl_start does on UnWeb MCP Server

AI agents invoke crawl_start to trigger actions in UnWeb 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.

WHY IT NEEDS A POLICY

Why crawl_start needs a policy

This tool initiates an active crawling operation that fetches multiple external web pages and processes them. It triggers an ongoing external operation (a crawl job) whose scope and effects depend on the URL argument provided. This goes beyond a simple read/fetch of a single resource — it spawns a background job that can touch many pages, consuming bandwidth and external server resources.

From the tool's definition 'Start a web crawl job. Crawls pages under the given URL path and converts them to markdown. Returns a job ID for tracking progress'

Attacks that exploit this kind of access

Recommended rule Ready

Verdict

crawl_start Rate-limited

crawl_start 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": {
    "crawl_start": {
      "limits": [
        {
          "counter": "crawl_start_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 UnWeb MCP Server, so you can run agents against it safely.

FAQ

Questions about crawl_start

What does the crawl_start tool do? +

Start a web crawl job. Crawls pages under the given URL path and converts them to markdown. Returns a job ID for tracking progress with crawl_status. It is categorised as a Execute tool in the UnWeb 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 crawl_start? +

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

What risk level is crawl_start? +

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

Can I rate-limit crawl_start? +

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

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

crawl_start is provided by the UnWeb MCP Server MCP server (mbsoft-systems/unweb-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.