Vss Deploy Video Embedding logo

Vss Deploy Video Embedding

OrganizationPopular
NVIDIA-AI-Blueprints
vss-deploy-video-embedding

Use this skill when deploying, operating, integrating, or customizing the VSS RT-Embed Video Embedding microservice. Covers standalone Docker Compose deployment, the `/v1` REST API for text/video embeddings and live streams, Redis/Kafka/OTel integration, troubleshooting, and bring-your-own-model (BYOM) custom embedding backends, with VideoPrism as an example. Do not use for RT-CV, RT-VLM, VSS Agent, or general VSS deployment work that does not include RT-Embed.

Overview

PublisherNVIDIA-AI-Blueprints
Repositoryvideo-search-and-summarization
Skill namevss-deploy-video-embedding
Stars
1.9K
Forks
393
Bundled files
12
LicenseApache-2.0
Links
Skill sourceRepositoryHomepage
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 12 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by NVIDIA-AI-Blueprints on GitHub. Read the source before you install it.

Installation

Install the Vss Deploy Video Embedding AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/NVIDIA-AI-Blueprints/video-search-and-summarization.git /tmp/video-search-and-summarization
mkdir -p .claude/skills
cp -r /tmp/video-search-and-summarization/skills/deployment/vss-deploy-video-embedding .claude/skills/vss-deploy-video-embedding
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Vss Deploy Video Embedding in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Vss Deploy Video Embedding on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Vss Deploy Video Embedding is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

VSS Video Embedding (RT-Embed)

Use this skill for the RT-Embed video embedding microservice, including the standard Cosmos-Embed1 deployment path and custom/BYOM embedding model work.

Trigger phrases: vss-deploy-video-embedding, RT-Embed, rtvi-embed, video embedding service, Cosmos-Embed1, embed live stream, embed video file, generate video embeddings, text embedding for video search, RT-Embed BYOM, VideoPrism embed, custom embed model, MODEL_IMPLEMENTATION_PATH, MODEL_REPOSITORY_SCRIPT_PATH, bring your own embedding model.

Do not use this skill for RT-CV, RT-VLM, VSS Agent, or general VSS deployment work unless the request deploys, operates, integrates, or customizes RT-Embed.

Service Snapshot

  • Skill: vss-deploy-video-embedding.
  • Legacy 3.1 name: RT-Embed.
  • Compose service: rtvi-embed.
  • Container name: vss-rtvi-embed.
  • Image: ghcr.io/nvidia-ai-blueprints/vss/vss-rt-embed (override with VSS_RT_EMBED_IMAGE).
  • Default tag: develop-latest (override with VSS_RT_EMBED_TAG; use develop-latest-sbsa for an SBSA/DGX Spark host).
  • Profile: rtvi-embed.
  • Container port: 8000 (host-side ${RTVI_EMBED_PORT}).
  • Default model: cosmos-embed1-448p from nvidia/Cosmos-Embed1-448p.
  • BYOM loader variables: MODEL_PATH, MODEL_IMPLEMENTATION_PATH, MODEL_REPOSITORY_SCRIPT_PATH.
  • Health endpoint: GET /v1/ready.
  • Healthcheck startup grace: 1200s (20 minutes) on first boot.

Route First

Choose one primary path before acting. Load the linked reference and follow it; do not duplicate full workflows from this top-level file.

User intentUse this path
Deploy, size, upgrade, roll back, or tear down standalone RT-Embed with the default Cosmos-Embed1 modelreferences/deploy-vss-deploy-video-embedding.md
Call RT-Embed APIs for files, text/video embeddings, live streams, model listing, health, metrics, metadata, or manifestsreferences/rest-api.md
Wire RT-Embed into another service or deployment with Redis, Kafka, OpenTelemetry, auth, storage, or env var mappingreferences/integrate-vss-deploy-video-embedding.md and references/environment.md
Use decoded-frame IPC from a compatible RTVI CV producerreferences/environment.md#decoded-frame-ipc and references/integrate-vss-deploy-video-embedding.md
Add, wire, or validate a custom/BYOM embedding backend, with VideoPrism as an examplereferences/byom-custom-model.md
Debug readiness, model/cache startup, permissions, Redis/Kafka reachability, API failures, or observabilityreferences/troubleshooting.md

Selection rules:

  • If asked which skill handles default Cosmos-Embed1 RT-Embed deployment, answer: use vss-deploy-video-embedding; this is the default deployment path, not the BYOM/custom-model path.
  • For normal RT-Embed or Cosmos-Embed1 deployment, use the deployment reference. In the answer, explicitly say that this is the default RT-Embed deployment path. Also explicitly distinguish it from BYOM/custom model integration: BYOM is only for adding or validating non-default custom embedding backends such as VideoPrism, and is not needed for the default Cosmos-Embed1 model.
  • For BYOM, custom embedding models, VideoPrism examples, or model implementation path questions, use the BYOM reference first, then deployment/API references only as needed.
  • For direct endpoint calls, use the API reference and reuse deployment context only when the service is not already running.
  • Decoded-frame IPC requires a compatible RTVI CV producer on the same host, a shared socket directory accessible to UID/GID 1001, and a matching camera ID. The consumer uses the fixed /run/rtvi-ipc/nvds_ipc_{camera_id}.sock socket contract. IPC camera, sensor, and stream IDs must be non-empty and contain only ASCII letters, digits, ., _, and -; standard UUIDs are valid. It applies only to live RTSP processing.
  • If the request mixes deployment and BYOM, load BYOM first to establish model path requirements, then use the deployment reference to run the service.

Operating Rules

  • Do not deploy a full VSS profile for standalone RT-Embed. Work from deploy/docker/services/rtvi/rtvi-embed unless the user explicitly asks for a profile deployment.
  • Never let sudo prompt interactively. Prefer plain docker; otherwise use sudo -n docker and stop with the exact manual command if passwordless sudo is unavailable.
  • Do not expose full values of NGC_API_KEY, HF_TOKEN, bearer tokens, or model repository credentials in prompts, logs, or final answers.
  • Do not shorten the start_period: 1200s healthcheck during first boot. Cosmos model download and Triton model repository generation can take up to 20 minutes.
  • In standalone mode, disable missing peers with MESSAGE_BUS=, ERROR_BUS=, and ENABLE_REDIS_ERROR_MESSAGES=false unless the corresponding Kafka or Redis service is started and reachable.
  • For BYOM models that are video-only, require an explicit text endpoint decision: either a compatible text encoder in the same embedding space or a clear 4xx response for /v1/generate_text_embeddings.

Quick Reference

References

FileWhen to read
references/README.mdTable of contents for all reference files.
references/deploy-vss-deploy-video-embedding.mdDeployment reference: image, GPU, storage, startup, prerequisites, known issues.
references/rest-api.mdFull REST endpoint catalog with worked curl examples for file uploads, video/text embeddings, live streams, and health/metrics.
references/integrate-vss-deploy-video-embedding.mdIntegration reference: peers, inputs/outputs, env vars, network, example Compose snippet.
references/environment.mdComplete environment-variable matrix, including host-to-container renames and secret-sensitive variables.
references/byom-custom-model.mdBYOM reference: custom model contract, path overrides, Docker/Helm wiring, and VideoPrism example validation checklist.
references/troubleshooting.mdOperational diagnostics for startup, model/cache, runtime, and observability issues.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Vss Deploy Video Embedding AI skill do?

Use this skill when deploying, operating, integrating, or customizing the VSS RT-Embed Video Embedding microservice. Covers standalone Docker Compose deployment, the `/v1` REST API for text/video embeddings and live streams, Redis/Kafka/OTel integration, troubleshooting, and bring-your-own-model (BYOM) custom embedding backends, with VideoPrism as an example. Do not use for RT-CV, RT-VLM, VSS Agent, or general VSS deployment work that does not include RT-Embed.

Why use Vss Deploy Video Embedding on TypingMind?

Because you install it once and use it with any model. Vss Deploy Video Embedding is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Vss Deploy Video Embedding in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/NVIDIA-AI-Blueprints/video-search-and-summarization/tree/develop/skills/deployment/vss-deploy-video-embedding. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Vss Deploy Video Embedding?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Vss Deploy Video Embedding?

As many as you like. As long as a model supports skills, you can use Vss Deploy Video Embedding with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Vss Deploy Video Embedding AI skill free?

Yes. It is published on GitHub by NVIDIA-AI-Blueprints under the Apache-2.0 license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

Related AI Skills

View all
garrytan logo

Setup Deploy

garrytan
CommunityPopular

Configure deployment settings for /land-and-deploy.

19.9K
133.5K
1 files
View
JuliusBrussee logo

Caveman Compress

JuliusBrussee
CommunityPopular

Compress a memory file such as CLAUDE.md or a todo list into caveman format to save input tokens, keeping a readable backup. Trigger: /caveman-compress.

6.2K
106.3K
7 files
View
JuliusBrussee logo

Caveman Discover

JuliusBrussee
CommunityPopular

Find and label every LLM workflow in the repository so Caveman Cloud groups spend by workflow instead of one bucket. Use for "discover workflows" or breaking LLM spend down by workflow.

6.2K
106.3K
Instructions
View
JuliusBrussee logo

Caveman Manage

JuliusBrussee
CommunityPopular

Inspect Caveman Cloud's experiment lifecycle and block unsafe execution. Use when asked to start, approve, cancel, promote or roll back a Caveman experiment.

6.2K
106.3K
Instructions
View

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇