AI agents call fetch_html to retrieve information from MCP NPX Fetch without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves web content and formats it; it performs no side effects, does not execute code, does not modify data, and does not delete anything. It is a straightforward read operation analogous to a GET request or web scraping. Severity is low because the blast radius of misuse is minimal—an agent might retrieve sensitive public web content, but cannot cause destructive or financial harm with this tool alone.
From the tool's definition Tool description explicitly states 'Fetch a website and return the content as HTML' — a retrieval operation with no modification, deletion, or code execution.
Documented attack patterns abuse exactly the kind of access fetch_html gives an agent:
PolicyLayer is an MCP gateway — it sits between your AI agents and MCP NPX Fetch, and nothing reaches the server without passing your rules. This is the rule we recommend for fetch_html:
{
"version": "1",
"default": "deny",
"tools": {
"fetch_html": {}
}
}fetch_html is read-only, so it stays allowed — but everything else on the server is denied unless you say otherwise.
Free to start. No card required.
Fetch a website and return the content as HTML. It is categorised as a Read tool in the MCP NPX Fetch MCP Server, which means it retrieves data without modifying state.
Register the MCP NPX Fetch MCP server in PolicyLayer and add a rule for fetch_html: 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 NPX Fetch. Nothing to install.
fetch_html is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the fetch_html 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.
Set action: deny in the PolicyLayer policy for fetch_html. 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.
fetch_html is provided by the MCP NPX Fetch MCP server (tokenizin-agency/mcp-npx-fetch). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Deterministic rules across all 4 MCP NPX Fetch tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.
Free to start. No card required.
4 MCP NPX Fetch tools catalogued and risk-classified — across an index of 42,500+ MCP servers.