Create a new TCP/UDP stream proxy
AI agents use create_stream to create or update resources in Nginx Proxy Manager MCP — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Nginx Proxy Manager MCP environment.
This tool creates new infrastructure configuration (a stream proxy) which modifies the system state and proxy routing behavior. It is classified as Write rather than Execute because it is a configuration creation action that is reversible.
From the tool's definition Tool creates a new TCP/UDP stream proxy configuration in Nginx Proxy Manager. The action is reversible (can be deleted), modifies the proxy infrastructure by adding a new routing rule, and represents a write operation that changes system configuration.
Attacks that exploit this kind of access
Create a new TCP/UDP stream proxy. It is categorised as a Write tool in the Nginx Proxy Manager MCP MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Nginx Proxy Manager MCP server in PolicyLayer and add a rule for create_stream: 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 Nginx Proxy Manager MCP. Nothing to install.
create_stream is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.
Yes. Add a rate_limit block to the create_stream 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 create_stream. 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.
create_stream is provided by the Nginx Proxy Manager MCP server (verybigsad/nginx-proxy-manager-mcp). 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 →