← All skills
vss-build-vision-ai
67 / 100 · Review before installing
Check before installing. Some of what it can do is expected: a coding skill usually runs build and test commands.
By NVIDIA/skills (GitHub) · coding · 3,933 stars · scanned at commit 2630133
Add agent-ready vision capabilities — dense captioning, detection, search, alerting, summarization — to an agent or application through a customizable, self-contained vision stack built on the NVIDIA VSS Blueprint. Use this skill when a developer or agent wants to give their app vision: pick capabilities via guided intake ("build a vision agent", "add vision capabilities") or describe them in natural language ("create a profile for streaming dense captioning", "add agentic search to my base dep…
Works with any agent that supports SKILL.md skills.
What this skill can do
Runs commands on your computer
This is disclosed by the author in the skill's SKILL.md or README. Expected for a coding skill.
Found in 42 places.
Show 17 more
- skills/vss-build-vision-ai/config/skillspector-baseline.yml, line 135
subprocess.run(env=...) so exported tags cannot skew the result. Nothing - skills/vss-build-vision-ai/evals/vdr_1_quickstart_vision_agent.json, line 45
"`sh -c 'i=1; while [ \"$i\" -le 24 ]; do curl -sf --max-time 10 http://localhost:7777/ >/dev/null 2>&1 && exit 0; sleep 5; i=$((i+1)); done; exit 1'` returns e - skills/vss-build-vision-ai/evals/vdr_1_quickstart_vision_agent.json, line 46
"`sh -c 'cd {{repo_root}} && uv run --project libs/vss vss --version >/dev/null 2>&1; i=1; while [ \"$i\" -le 24 ]; do uv run --project libs/vss vss configure - - skills/vss-build-vision-ai/evals/vdr_1_quickstart_vision_agent.json, line 47
"`sh -c 'i=1; while [ \"$i\" -le 24 ]; do docker ps --format \"{{.Names}}\" | grep -qx vss-agent-ui && exit 0; sleep 5; i=$((i+1)); done; exit 1'` returns exit - skills/vss-build-vision-ai/evals/vdr_1_quickstart_vision_agent.json, line 48
"`sh -c 'i=1; while [ \"$i\" -le 24 ]; do docker ps --format \"{{.Names}}\" | grep -qx redis && exit 0; sleep 5; i=$((i+1)); done; exit 1'` returns exit 0 (shar - skills/vss-build-vision-ai/evals/vdr_1_quickstart_vision_agent.json, line 50
"`sh -c 'i=1; while [ \"$i\" -le 24 ]; do docker ps --format \"{{.Names}}\" | grep -qx vss-rtvi-vlm && exit 0; sleep 5; i=$((i+1)); done; exit 1'` returns exit - skills/vss-build-vision-ai/evals/vdr_2_add_alerting_summarization.json, line 23
"`sh -c 'cd {{repo_root}} && uv run --project libs/vss vss --version >/dev/null 2>&1; i=1; while [ \"$i\" -le 24 ]; do uv run --project libs/vss vss configure - - skills/vss-build-vision-ai/evals/vdr_2_add_alerting_summarization.json, line 39
"`sh -c 'cd {{repo_root}} && uv run --project libs/vss vss --version >/dev/null 2>&1; i=1; while [ \"$i\" -le 24 ]; do uv run --project libs/vss vss configure - - skills/vss-build-vision-ai/evals/vdr_2_add_alerting_summarization.json, line 40
"`sh -c 'i=1; while [ \"$i\" -le 24 ]; do docker ps --format \"{{.Names}}\" | grep -qx vss-alert-bridge && exit 0; sleep 5; i=$((i+1)); done; exit 1'` returns e - skills/vss-build-vision-ai/evals/vdr_2_add_alerting_summarization.json, line 41
"`sh -c 'i=1; while [ \"$i\" -le 36 ]; do s=$(docker inspect -f \"{{if .State.Health}}{{.State.Health.Status}}{{else}}{{.State.Status}}{{end}}\" vss-lvs 2>/dev/ - skills/vss-build-vision-ai/evals/vdr_2_add_alerting_summarization.json, line 42
"`sh -c 'i=1; while [ \"$i\" -le 24 ]; do docker ps --format \"{{.Names}}\" | grep -qx vss-video-analytics-api && exit 0; sleep 5; i=$((i+1)); done; exit 1'` re - skills/vss-build-vision-ai/evals/vdr_2_add_alerting_summarization.json, line 43
"`sh -c 'i=1; while [ \"$i\" -le 24 ]; do docker ps --format \"{{.Names}}\" | grep -qx vss-rtvi-vlm && exit 0; sleep 5; i=$((i+1)); done; exit 1'` returns exit - skills/vss-build-vision-ai/references/prerequisites.md, line 505
sudo bash -c 'cat > /etc/docker/daemon.json << EOF - skills/vss-build-vision-ai/scripts/stage_vss_src.py, line 50
r = subprocess.run( - skills/vss-build-vision-ai/scripts/stage_vss_src.py, line 69
except subprocess.CalledProcessError: - skills/vss-build-vision-ai/scripts/stage_vss_src.py, line 139
except (subprocess.CalledProcessError, OSError) as exc: - skills/vss-build-vision-ai/scripts/stage_vss_src.py, line 163
except (subprocess.CalledProcessError, OSError, FileNotFoundError) as exc:
and 22 more places
Sends data to these websites: astral.sh, 169.254.169.254, nvidia.github.io, ${host_ip}
The author does not mention this.
Found in 75 places.
Show 17 more
- skills/vss-build-vision-ai/references/brev.md, line 222
curl -sfI http://localhost:7777/ | head -1 - skills/vss-build-vision-ai/references/brev.md, line 240
| `curl "$(brev_origin 7777)"` → 502 | HAProxy container (`vss-haproxy-ingress`) is down — `docker logs vss-haproxy-ingress` | - skills/vss-build-vision-ai/references/brev.md, line 241
| `curl "$(brev_origin 7777)"` → Cloudflare Access login page forever | User hasn't been granted access in the Brev org; not a deploy issue | - skills/vss-build-vision-ai/references/credentials.md, line 156
curl -s -o /dev/null -w '%{http_code}\n' --connect-timeout 5 --max-time 15 \ - skills/vss-build-vision-ai/references/deployment_resolution.md, line 70
if _models=$(curl -sf --max-time 5 "${_candidate}/models") \ - skills/vss-build-vision-ai/references/deployment_resolution.md, line 82
answers `curl -sf` but yields no `.data[0].id`, so that candidate is rejected - skills/vss-build-vision-ai/references/edge.md, line 203
curl -sf http://localhost:30081/v1/health/ready && echo "LLM ready" - skills/vss-build-vision-ai/references/edge.md, line 204
curl -s http://localhost:30081/v1/models | jq -r '.data[].id' - skills/vss-build-vision-ai/references/ngc.md, line 35
curl -sLo /tmp/ngccli.zip \ - skills/vss-build-vision-ai/references/ngc.md, line 47
curl -sLo /tmp/ngccli.zip \ - skills/vss-build-vision-ai/references/prerequisites.md, line 31
curl -LsSf https://astral.sh/uv/install.sh | sh - skills/vss-build-vision-ai/references/prerequisites.md, line 267
| Cloud VM (AWS/GCP/Azure) | the **public/elastic IP** — **not on the NIC** (provider NAT, so `ip route`/`ip addr` can't see it). Read from instance metadata, e - skills/vss-build-vision-ai/references/prerequisites.md, line 545
curl -fsSL https://nvidia.github.io/libnvidia-container/gpgkey \ - skills/vss-build-vision-ai/references/prerequisites.md, line 547
curl -s -L https://nvidia.github.io/libnvidia-container/stable/deb/nvidia-container-toolkit.list \ - skills/vss-build-vision-ai/references/profiles/alerts.md, line 110
curl -sf "http://${HOST_IP}:8000/health" - skills/vss-build-vision-ai/references/profiles/alerts.md, line 113
curl -sf "http://${HOST_IP}:${VIDEO_ANALYTICS_API_HOST_PORT:-8081}/livez" - skills/vss-build-vision-ai/references/profiles/alerts.md, line 116
curl -sf "http://${HOST_IP}:${ALERT_BRIDGE_HOST_PORT:-9080}/health"
and 55 more places
Changes files outside its own folder
The author does not mention this.
Found in 14 places.
Show 11 more
- skills/vss-build-vision-ai/references/ngc.md, line 37
sudo mkdir -p /usr/local/lib - skills/vss-build-vision-ai/references/ngc.md, line 40
sudo ln -sfn /usr/local/lib/ngc-cli/ngc /usr/local/bin/ngc - skills/vss-build-vision-ai/references/prerequisites.md, line 189
cat <<'EOF' | sudo tee /etc/sysctl.d/99-vss.conf - skills/vss-build-vision-ai/references/prerequisites.md, line 475
test -f /etc/docker/daemon.json || echo '{}' | sudo tee /etc/docker/daemon.json >/dev/null - skills/vss-build-vision-ai/references/prerequisites.md, line 476
sudo cp /etc/docker/daemon.json /etc/docker/daemon.json.bak - skills/vss-build-vision-ai/references/prerequisites.md, line 481
&& sudo install -m 0644 /tmp/daemon.json.new /etc/docker/daemon.json \ - skills/vss-build-vision-ai/references/prerequisites.md, line 499
sudo cp /etc/docker/daemon.json /etc/docker/daemon.json.bak 2>/dev/null || true - skills/vss-build-vision-ai/references/prerequisites.md, line 505
sudo bash -c 'cat > /etc/docker/daemon.json << EOF - skills/vss-build-vision-ai/references/prerequisites.md, line 520
/etc/docker/daemon.json.bak | sudo tee /etc/docker/daemon.json >/dev/null - skills/vss-build-vision-ai/references/prerequisites.md, line 549
| sudo tee /etc/apt/sources.list.d/nvidia-container-toolkit.list - skills/vss-build-vision-ai/scripts/tests/test_sync_skills.py, line 283
body = re.search(r"printf '(#!/bin/sh\\n.*?)' > /usr/local/bin/vss-hermes-sync", dockerfile).group(1)
Downloads and installs other software
The author does not mention this.
Found in 7 places.
Show 4 more
Safer alternatives
Low-risk skills in the same category (coding) that look related. Check that one does what you need.
Install
Paste this into your terminal. It downloads exactly the version we scanned, not whatever is newest.
if [ -e ~/.agents/skills/'vss-build-vision-ai' ] || [ -L ~/.agents/skills/'vss-build-vision-ai' ]; then echo 'A skill folder named vss-build-vision-ai already exists — not replacing it' >&2; false; else d=$(mktemp -d) && git clone --quiet --filter=blob:none --no-checkout 'https://github.com/NVIDIA/skills' "$d" && git -C "$d" checkout --quiet '2630133f22e357f2005886567a39eddcc1b520e2' -- 'skills/vss-build-vision-ai' && mkdir -p ~/.agents/skills && cp -R "$d"/'skills/vss-build-vision-ai' ~/.agents/skills/'vss-build-vision-ai' && rm -rf "$d"; fi
Codex reads ~/.agents/skills directly. For Claude Code, Cursor, Copilot, Gemini CLI, Antigravity or OpenCode, also link it. For Claude Code: ln -s ~/.agents/skills/'vss-build-vision-ai' ~/.claude/skills/'vss-build-vision-ai' (other agents: ~/.cursor/skills, ~/.copilot/skills, ~/.gemini/skills, ~/.gemini/config/skills, ~/.config/opencode/skills). Or install through the finder skill, which does both.
Badge for your README
The badge for NVIDIA/skills shows the score of its lowest-scoring skill.
[](https://zalavijaysinh.github.io/skilltrust/repos/NVIDIA/skills/)
Score history
| Date | Version | Score |
|---|
| 2026-11-11 | 2630133 | 67 |
| 2026-11-08 | fce3cca | 67 |
| 2026-11-07 | fce3cca | 57 |