Command

Meter Command Handbook

Your network. At your Command.

For decades, managing enterprise networks has meant working around the tools available, not necessarily with the tools you want.

First came the command-line interface (CLI). Fast, powerful, and extremely precise—but also cryptic, intimidating, and inaccessible to most. Then came cloud dashboards. They brought visibility and ease of use but took away some of the control and customization that network engineers needed. Today’s software, especially in networking, often feels more rigid than ever: locked into specific workflows, dependent on structured UI flows, and lacking the flexibility to reflect how teams actually work.

Command changes that.

Send

Meter Command is a new kind of interface: a generative UI that lets you talk to your network. Ask a question, and Command will fetch live data. Describe a task, and Command will perform it. Request a dashboard, and Command will build one from scratch—on the fly and tailored to you.

Command is what happens when a software interface understands your intent and has access to your actual infrastructure. It takes your plain language input and builds precisely what you need, whether that’s a dashboard, a configuration change, or an answer. It does this by leveraging the full Meter stack: our hardware, telemetry, security, cloud services, and real-time data pipelines.

The architecture behind Command

At the heart of Command is a custom intelligence architecture designed to translate user intent into action, accurately and instantly. This system is built on three foundational stages:

What three clients experienced the most disruptions in the last 24 hours?
Research
  • Planning: When you ask Command a question—say, “What three clients experienced the most disruptions in the last 24 hours?”—the system breaks it down. What data does this require? What devices or services are involved? What filters or time ranges are implied? Command plans a sequence of internal steps that map your request to the backend.

  • Execution: Command then runs those steps in parallel. It pulls live data from multiple sources, filters it based on your request, and organizes it for display. This all happens in milliseconds and allows for complex queries to be answered as fast as a web search.

  • Synthesis: Finally, Command composes a result. That could be a chart, a dashboard, a configuration form, or a concise summary. If the request is actionable—like "Turn on port 19"—it builds an interface to confirm and execute that change, complete with context and safeguards.

@MDF-C01
turn on port 19
Physical Interface UUID
Enable port
All available fieldsSelect any additional fields you would like to include:
Enable port
Submit
This operation updates the physical interface settings of a network device. It may temporarily affect network performance or connectivity for devices connected to this interface. Ensure you understand the potential impact on your network before proceeding.

Because Command is trained on the way Meter's systems actually work—from frontend logic to backend telemetry—it doesn't just answer questions. It generates fully functional software with accuracy and precision.

Privacy and speed by design

Command is built for performance, but never at the expense of privacy.

Privacy

  • All data stays within Meter’s secure infrastructure.
  • No customer data is shared with third-party model providers.
  • Processing follows strict SOC 2 Type 2 standards.
  • Encryption is applied to all data at rest and in transit.
  • Role-based access controls are enforced throughout the stack.

Speed

  • Most responses return in under a few seconds.
  • Parallel execution architecture dramatically reduces latency.
  • Custom model serving infrastructure minimizes overhead.
  • Dashboards and components are generated on demand and update in real time.
  • Command is fast enough to support your workflow and secure enough to trust with your infrastructure.

How to use Meter Command

  1. Type your request: Start by asking Command a question or giving it a task. When referring to a device or object, you will need to tag it through the @ functionality to select a device instead of referring to it only by un-tagged label. You don’t need to use technical syntax, though it supports that too.
Primary
Send

A list of 28 different kinds of objects (see below), ranging from Networks, Clients, Switches, PDUs, WAN Interfaces, among others, will appear when you type @.

organizational
physical
policy
  1. Review and act: Command will return a live view or pre-filled form based on your input. You can:

    • Inspect real-time metrics.
    • Confirm and execute configuration changes.
    • Save views as dashboards.
    • Set alerts or filters.
    • Share outputs with your team.
  2. Build and share dashboards: Every query can become a dashboard component, simply drag it from the chatbox to the infinite canvas on the right. Group components together by region, use case, or role. You can even create shared workspaces and notebooks for team collaboration.

Show me a utilization chart for
@AP02
Primary
Send
AP utilization
AP02
24h
5G Band
2.4G Band

Prompt library

The following examples are grouped into six key categories. Each category includes prompts you can try—and a featured example to illustrate what's possible.

Configuration
Troubleshooting
Monitoring
Security & policies
Creative

A network that works the way you do

Command is built for the way you actually work, not the way legacy tools expect you to work. Whether you're deep in the stack or new to network operations, Command gives you real-time answers, smart automation, and full visibility—no certification required.

This is the interface for the next decade of infrastructure—one that’s flexible, fast, secure, and human. It’s your network, at your Command.

For more information, demos, or support, visit meter.com/command or email hello@meter.com.