skills.rest
DEEPX-AI
DEEPX-AIOFFICIAL
Explore all skills by this author →
WebsiteGitHub

dx-agent-stream-build-mqtt-kafka

Builds GStreamer pipelines that publish inference results to Kafka or MQTT brokers.

3|5|Updated Jan 21, 2025
Search Tags:#pipeline#kafka#object detection#mqtt#message broker#gstreamer#deepx
One-click install
npx skills add https://github.com/DEEPX-AI/dx_stream --skill dx-agent-stream-build-mqtt-kafka-deepx-ai
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: dx-agent-stream-build-mqtt-kafka
Source: https://github.com/DEEPX-AI/dx_stream/tree/main/.deepx/skills/dx-agent-stream-build-mqtt-kafka
Command: npx skills add https://github.com/DEEPX-AI/dx_stream --skill dx-agent-stream-build-mqtt-kafka-deepx-ai

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires kafka-python, paho-mqtt.

What problem does it solve? Connecting a DEEPX NPU vision inference pipeline to an external message broker requires correctly wiring DxMsgConv and DxMsgBroker elements, writing producer scripts, and building consumer clients, which is error-prone when done manually. ## Core Features & Use Cases - Kafka Producer Pipelines: Generates gst-launch-1.0 scripts that stream detection results as JSON to a Kafka topic, plus a Python Kafka consumer. - MQTT Publisher Pipelines: Generates MQTT publisher pipelines with a Python paho-mqtt subscriber for receiving inference messages. - Isolated Output Sessions: Creates all generated applications under timestamped dx-agent-dev/ session directories with session.json metadata and run scripts. - Use Case: A developer wants to stream YOLOv5 object detection results from a video file into a Kafka topic so a downstream analytics service can consume structured JSON detections in real time. ## Quick Start Ask the agent to build a broker pipeline that publishes YOLOv5 detection results to a Kafka topic named detections on localhost:9092.

Frequently Asked Questions about dx-agent-stream-build-mqtt-kafka

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I publish GStreamer inference results to Kafka?▼

Build a pipeline ending with dxmsgconv followed by dxmsgbroker configured with broker-name=kafka, a connection address, and a topic. DxMsgConv converts detection metadata to JSON and DxMsgBroker publishes it to the Kafka broker.

How do I send object detection results to an MQTT broker?▼

Use a pipeline with dxpreprocess, dxinfer, dxpostprocess, dxmsgconv, and dxmsgbroker with broker-name=mqtt and your broker address and topic. A paho-mqtt Python subscriber can then receive the JSON detection messages.

What is the difference between DxMsgConv and DxMsgBroker?▼

DxMsgConv converts inference metadata into JSON messages and must precede DxMsgBroker in the pipeline. DxMsgBroker is a terminal element with no source pad that transmits those messages to the configured Kafka or MQTT broker.

What Python dependencies are needed for the Kafka and MQTT consumers?▼

The Kafka consumer requires the kafka-python package installed via pip. The MQTT subscriber requires paho-mqtt, and a running broker such as Mosquitto or a Kafka container must be reachable at the configured address.

Where are generated broker pipeline files created?▼

All generated applications are created under dx-agent-dev/ in a timestamped session directory containing run scripts, consumer code, configuration files, and a session.json metadata file. Production directories are only used when explicitly requested.

Related Skills

Explore More Software Engineering Skills→
eyepop-aieyepop-ai
2

eyepop-sdk

Deploy computer vision inference pipelines using the EyePop.ai Python SDK.

Official
Intermediate
mthang1801mthang1801

microservices

Configure RabbitMQ or Kafka brokers and shared patterns for NestJS microservices.

Community
Advanced
NVIDIANVIDIA
3.2k

deepstream-dev

Builds NVIDIA DeepStream video analytics pipelines using the Python pyservicemaker API.

Official
Advanced
nexu-ionexu-io
8.2k

mockup-device-3d

Render HTML content on 3D iPhone and MacBook device screens using CSS 3D transforms.

Official
Advanced