skills.rest
atomicrajat
atomicrajatCOMMUNITY
Explore all skills by this author →
GitHub

vss-setup-behavior-analytics

Deploys the vss-behavior-analytics container standalone with chosen entrypoint, config, and calibration.

2|Updated Aug 20, 2026
Search Tags:#deployment#calibration#docker compose#kafka#video analytics#behavior analytics#nvidia vss
One-click install
npx skills add https://github.com/atomicrajat/industry_safety_monitoring_system --skill vss-setup-behavior-analytics-atomicrajat
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: vss-setup-behavior-analytics
Source: https://github.com/atomicrajat/industry_safety_monitoring_system/tree/main/.claude/skills/vss-setup-behavior-analytics
Command: npx skills add https://github.com/atomicrajat/industry_safety_monitoring_system --skill vss-setup-behavior-analytics-atomicrajat

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Deploying the NVIDIA VSS behavior-analytics service outside the full warehouse blueprint stack requires knowing which entrypoint to pick, which config and calibration files to mount, and how to wire dynamic runtime updates over Kafka. This Skill guides that standalone deployment end to end. ## Core Features & Use Cases - Standalone container deployment: Brings up only the vss-behavior-analytics service via Docker Compose, with entrypoint selection across analytics 2D, analytics 3D/mv3dt, dev_example, and fusion_search. - Config and calibration sourcing: Mounts profile-shipped or custom JSON configs and optional cartesian, geo, or image calibration files. - Dynamic runtime updates: Publishes upsert, upsert-all, and delete messages to the mdx-notification Kafka topic to change config or calibration without redeploying. - Use Case: An engineer wants to run behavior analytics against an existing Kafka broker without the full warehouse UI, agent, and perception stack, then push a new ROI rule to the running container. ## Quick Start Ask the agent to deploy vss-behavior-analytics standalone using the default Analytics 2D entrypoint and the warehouse 2D config.

Frequently Asked Questions about vss-setup-behavior-analytics

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

FAQPage Schema
How do I deploy vss-behavior-analytics standalone without the full warehouse stack?▼

Edit the compose file at deploy/docker/services/analytics/behavior-analytics/compose.yml to set the entrypoint, config volume, and optional calibration volume, then run docker compose up -d vss-behavior-analytics-base with VSS_APPS_DIR exported. The container starts even without a broker.

Which entrypoint should I choose for behavior analytics?▼

Use main_analytics_2d_app.py for 2D world-plane analytics, main_analytics_3d_app.py for 3D or mv3dt multi-view tracking, main_dev_example_app.py for FOV-count and restricted-area detection, or main_fusion_search_analytics_app.py for video-embedding downsampling with the search profile.

Can I update behavior-analytics config without restarting the container?▼

Yes, when a broker is reachable. Publish an upsert or upsert-all message to the mdx-notification Kafka topic with key behavior-analytics-config; the listener validates, persists, applies to all workers, and sends an ack. Calibration updates use the same topic with key calibration.

Why does the behavior-analytics container keep restarting?▼

The container restart loop happens when no Kafka, Redis, or MQTT broker is reachable. The client retries a bounded number of times, the app exits, and restart: always cycles it. Start a broker or override the restart policy to on-failure or no.

What happens if I skip the calibration file at startup?▼

The app starts with a DynamicCalibration wrapper behaving as image-plane calibration and waits for the first calibration notification on mdx-notification. Until one arrives, analytics run in pixel space with no real-world distances or ROI and tripwire firings.

Why does the Docker image pull fail with 401 or 403?▼

A 401 or 403 from nvcr.io means the NGC_CLI_API_KEY is missing, expired, or not scoped to NGC Catalog. Regenerate the key in your NGC account, export it, and run docker login with username $oauthtoken and the key via password-stdin.

Related Skills

Explore More Software Engineering Skills→
NVIDIANVIDIA
3.2k

vss-setup-behavior-analytics

Deploy the vss-behavior-analytics container standalone with chosen entrypoint, config, and calibration.

Official
Advanced
NVIDIA-AI-BlueprintsNVIDIA-AI-Blueprints
1.8k

vss-deploy-detection-tracking-3d

Deploy the RTVI-CV-3D MV3DT microservice for multi-camera 3D detection and tracking.

Official
Advanced
NVIDIA-AI-BlueprintsNVIDIA-AI-Blueprints
1.8k

vss-setup-behavior-analytics

Deploy the vss-behavior-analytics container standalone with chosen entrypoint, config, and calibration.

Official
Advanced
cuiosscuioss
4

tools-permission-doctor

Identifies Claude Code permission misconfigurations including duplicates, wildcards and anti-patterns without modifying files.

Official
Intermediate