vss-deploy-video-embedding

Deploy the VSS Video Embedding microservice with Docker Compose and GPU-accelerated Cosmos-Embed1.

Updated May 29, 2026
One-click install
npx skills add https://github.com/rblake2320/vigil --skill vss-deploy-video-embedding-rblake2320
Or copy as Structured Prompt for Agentβ–Ό
Please help me install this Agent Skill.
Skill: vss-deploy-video-embedding
Source: https://github.com/rblake2320/vigil/tree/main/.claude/skills/vss-deploy-video-embedding
Command: npx skills add https://github.com/rblake2320/vigil --skill vss-deploy-video-embedding-rblake2320

SYSTEM DOCUMENTATION & REQUIREMENTS

πŸ’‘ This Skill includes references (resource) components.

What problem does it solve?

Deploy, operate, and integrate the VSS Video Embedding microservice (RT-Embed) to generate video and text embeddings, manage live streams, and connect with Redis, Kafka, and OpenTelemetry in a production-ready pipeline.

Core Features & Use Cases

  • Docker Compose deployment with GPU-accelerated Cosmos-Embed1-448p model and Triton backend
  • REST API for file uploads, embeddings, and live-stream management plus health and metrics endpoints
  • Integration with Redis, Kafka, and OpenTelemetry for observability, messaging, and tracing
  • Use cases include video search, RAG pipelines, and real-time analytics

Quick Start

Bring up the RT-Embed service locally with docker compose using the bp_developer_search_2d profile and verify readiness via /v1/ready.

Frequently Asked Questions about vss-deploy-video-embedding

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

FAQPage Schema
How do I deploy a GPU-accelerated video embedding microservice using Docker Compose?β–Ό

Deploy GPU-accelerated video embedding microservices using Docker Compose by configuring the bp_developer_search_2d profile, mapping ports, setting environment variables, and verifying startup via the /v1/ready health check endpoint.

Can I integrate video embeddings with Redis, Kafka, and OpenTelemetry?β–Ό

Yes, you can integrate video embeddings with Redis, Kafka, and OpenTelemetry for production observability. The RT-Embed microservice connects with these platforms to enable messaging, caching, and distributed tracing within your pipeline.

Does the VSS Video Embedding service require an NVIDIA GPU for the Cosmos-Embed1 model?β–Ό

GPU acceleration is required for the Cosmos-Embed1-448p model. The RT-Embed deployment uses a GPU-enabled Docker Compose configuration with the Triton backend to process video and text vector embeddings.

What REST API endpoints are available for file uploads and live stream embeddings?β–Ό

The REST API provides endpoints for file uploads, embedding generation, and live-stream management. It also includes dedicated health and metrics endpoints to monitor the operational status of the RT-Embed service.

Why is my RT-Embed video embedding deployment not passing startup health checks?β–Ό

Your RT-Embed deployment may fail health checks due to incorrect environment variables, missing model downloads, or improper port mappings. The service provides troubleshooting and rollback guidance to verify readiness via the /v1/ready endpoint.