AI agents call cf_tunnels to retrieve information from Access without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves infrastructure information (tunnel inventory, IDs, names, status) from a Cloudflare account without side effects. It is a read operation. However, severity is elevated to 'high' because: (1) the server handles credential storage and API proxying for AI agents with a single Bearer token, creating high blast radius if compromised; (2) enumeration of tunnels exposes network infrastructure topology…
From the tool's definition Tool description states 'List all Cloudflare Tunnels in an account with their IDs, names, and connection status' — explicitly a read/list operation with no modification or deletion capability.
Attacks that exploit this kind of access
List all Cloudflare Tunnels in an account with their IDs, names, and connection status. Use this to find tunnel IDs for cf_tunnel_config. It is categorised as a Read tool in the Access MCP Server, which means it retrieves data without modifying state.
Register the Access MCP server in PolicyLayer and add a rule for cf_tunnels: 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 Access. Nothing to install.
cf_tunnels 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 cf_tunnels 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 cf_tunnels. 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.
cf_tunnels is provided by the Access MCP server (scottpedia0/access). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →