# FAQ

## Frequently asked questions

### What is Device Agent?

Device Agent is an MQTT-powered AI Assistant that turns any IoT device into an AI agent in minutes.
Describe a device or upload a hardware definition to generate its specification, SDK, simulator, and IM/A2A integration.

### What kinds of devices is Device Agent for?

It fits devices and systems with clear state, commands, or events, such as sensors, cameras, door locks, air conditioners, robots, industrial gateways, lab equipment, edge services, and existing IoT systems.

### How is Device Agent different from chatbots, Claude Code, Codex, and OpenClaw?

Chatbots are message-centric. Claude Code and Codex are repository- and development-task-centric. OpenClaw is closer to general task automation. Device Agent is device-semantics-centric.

It does not treat devices as temporary tool APIs. It first uses DeviceSpec to define `commands`, `telemetry`, and `events`: what can be controlled, what is reported continuously, and which changes need handling. Based on that model, Device Agent owns device access, online state, command execution, and event handling, then exposes them through MQTT, SDKs, Web, voice, vision, IM, and EMQX A2A.

### What is DeviceSpec?

DeviceSpec is the device semantic definition for a device category. It describes basic device information, `commands`, `telemetry`, and `events`, so the Agent, SDKs, simulator, and A2A cards use the same capability model.

### Does one Device Agent manage only one device?

No. One Device Agent maps to one device category and one DeviceSpec. That device category can have many real devices connected under it. Each device has its own `deviceId`, online state, telemetry data, and event history, while sharing the same capability definition.

### Do I have to rebuild my existing device system?

Not necessarily. Existing firmware, gateways, or backend services can connect as long as they report state, respond to commands, and send events using the MQTT topics and payloads shown by Device Agent. New devices can start from SDK code generated by the console.

### Can Device Agent run automation?

Yes. Device Agent can trigger actions from conversations, telemetry, events, tool results, and external system state. Simple automation can run inside the Device Agent; cross-device or cross-system flows can be extended with tools, skills, or EMQX A2A.

### Does Device Agent support voice and vision?

Yes. Voice handles input and spoken replies. Vision adds images or scene context to device interactions. Like text and IM, both land on the same device semantics and runtime capabilities.

### What is A2A?

A2A is Device Agent's collaboration capability based on A2A over MQTT in EMQX 6.2.0 and later. When enabled, Device Agent publishes an A2A card from the DeviceSpec and maps `commands` into skills that other agents can discover and call.

EMQX handles card registration and discovery, online state, schema validation, and MQTT v5 request/reply communication. Other A2A agents can use that card to call the real device capabilities behind the current Device Agent.