NVIDIA AI Blueprint for Video Search and Summarization (VSS) is a reference architecture for applications that use AI to search, analyze, question, and summarize live or recorded video. It is intended for building video analytics agents with vision-language models, language models, retrieval-augmented generation, and NVIDIA NIM microservices. The catalogue entries provide coding-agent rules, skills, commands, agents, instructions, and settings associated with the project.
Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/NVIDIA-AI-Blueprints/video-search-and-summarizationWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/nvidia-ai-blueprints/video-search-and-summarization/vios-architecture)<a href="https://agentmods.dev/commands/nvidia-ai-blueprints/video-search-and-summarization/vios-architecture"><img src="https://agentmods.dev/badge/commands/nvidia-ai-blueprints/video-search-and-summarization/vios-architecture/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/nvidia-ai-blueprints/video-search-and-summarization/vios-architecture"><img src="https://agentmods.dev/badge/commands/nvidia-ai-blueprints/video-search-and-summarization/vios-architecture.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00042 | $0.05919 |
| Opus 5.5 | $0.00017 | $0.02368 |
| Sonnet 5.5 | $0.00008 | $0.01184 |
| Haiku 4.5 | $0.00004 | $0.00592 |
Grade A, and why
vios-architecture scanned grade A with 1 finding against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
actor U as User / UI / curl How it starts
The opening of the file, as written. The whole thing — 382 lines — stays where its author put it; the contents beside it link to each section on GitHub.
SPDX-FileCopyrightText: Copyright (c) 2026 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
SPDX-License-Identifier: MIT
Permission is hereby granted, free of charge, to any person obtaining a
copy of this software and associated documentation files (the "Software"),
to deal in the Software without restriction, including without limitation
the rights to use, copy, modify, merge, publish, distribute, sublicense,
and/or sell copies of the Software, and to permit persons to whom the
Software is furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in
all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
DEALINGS IN THE SOFTWARE.
VIOS Architecture — Stream-Processor + Sensor Deployment
This skill explains the stream-processor microservice (MS) + sensor-MS deployment of VIOS. It is the reference model used by the one-click deploy, the default in vios-deployment, and the topology BDD/sanity tests run against. Use it to build a mental model of what runs where, how requests are routed, and how state flows between services. It deliberately stays at the deployment / container / protocol level — for source-level details, grep the codebase.
1. Containers in the stack
| Container | Listens (host net) | Role |
|---|---|---|
vst-ingress (nginx) |
30888 |
Reverse proxy. Serves the React UI under /vst/, splits backend traffic between sensor-MS (30000) and the stream-processor data plane through Envoy (10000). |
sensor-ms |
30000 (HTTP), 8080 (Prom) |
Control plane. Owns sensor discovery, sensor add/delete, adaptor loading (vst_rtsp / onvif / milestone_onvif), stream monitoring, and the centralized DB. No recording, no RTSP server, no storage. |
streamprocessing-ms-N |
30001 (HTTP, pod 1), 30554 (RTSP, pod 1), 8087 (Prom) |
Data plane. One pod = one shard of streams. Owns recording, RTSP proxy server, live/replay WebRTC, clip generation, storage management, image capture. |
envoy-streamprocessing |
10000 |
Header-based router that fans streamid-keyed requests out to the right streamprocessing-ms-N pod. xDS pulls config from SDR. |
sdr-streamprocessing |
4003 |
Workload Distribution Manager. Watches Redis for sensor add/remove events, decides which pod owns a stream, attaches it to that pod, and writes the streamid→pod mapping to Redis for Envoy to read. |
centralizedb (postgres) |
5432 |
Shared metadata DB. Sensor list, recording manifests, configs. Both sensor-MS and every stream-processor pod open connections. |
redis-server |
6379 |
Message bus + routing table. SDR consumes the vst_events topic here; Envoy reads the streamid→pod hash here on every request. |
nvstreamer |
31000–31004 (HTTP, one per instance), 31554, 31564, 31574, 31584, 31594 (RTSP, spaced by 10) |
Separately deployed loopback RTSP source farm for testing / video-file-driven sensors. VIOS treats nvstreamer endpoints as a regular RTSP camera fleet. |
minio |
9000 / 9001 |
Optional object storage, gated by --profile minio. |
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago Changed · +46 lines 467bddcc90d4
- 1mo ago First seen · 336 lines · 42 tokens per session scan A ce728c3c5669
vios-architecture is a command published in the GitHub repository NVIDIA-AI-Blueprints/video-search-and-summarization (1,953 stars, last pushed today), licensed Apache-2.0. It adds 42 tokens to every session and 5,919 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other commands, from other repositories
create-docker-mcp-tunnel
Stand up an Anthropic MCP tunnel locally with Docker Compose so Claude can call a private MCP server (manual-credentials quickstart).
generate-manifest
Generate production-ready Kubernetes manifests from application configuration.
k8s-manifest
Generate production-ready Kubernetes manifests for the current application.
update-status
Update the status and progress of Linear tickets.
deployment
Basic production deployment setup for FastAPI applications.
rclone_serve
Serve a remote over a protocol.