菜单
Run the Live Video Captioning unit test suite (pytest in app/). Use when asked to test, verify, or check coverage of backend changes.
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/open-edge-platform/edge-ai-suites --skill lvc-test命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
Build an end-to-end, vertical-agnostic computer-vision analytics stack on Intel hardware from a single streamlined Docker Compose deployment: DL Streamer Pipeline Server plus MediaMTX/WebRTC, Coturn, Mosquitto, Node-RED, Grafana, and Nginx. It streams live annotated video over WebRTC and flows detection metadata DLSPS->MQTT->Node-RED->Grafana, with an optional SceneScape multi-camera spatial-analysis path. USE FOR standing up an object-detection, classification, counting, or zone-alerting pipeline for any vertical (smart city/ITS, retail, industrial, logistics, healthcare, or a custom OpenVINO/ONNX model) where only the model, class filter, alert rule, and dashboard change. Also USE FOR a lightweight **demo/PoC** single application (no full stack) — a simple DL Streamer pipeline (via the `dlstreamer-coding-agent` skill) or a simple OpenVINO inference app (guided by the OpenVINO 2026 docs) selected via the mode question. DO NOT USE FOR non-Intel or cloud-only deployments, Prometheus/OpenTelemetry metrics stack
Conversational orchestrator that turns a plain business objective into a working Intel Edge AI application. It OWNS the conversation: it asks only business questions (what outcome you want, your inputs, where it runs, your hardware) — never which framework, model, or device — then DISCOVERS the relevant skills from the open-edge-platform/skills catalog, proposes a plan, and only after you confirm builds the deliverable by DELEGATING to the right skill(s). USE FOR any "I want to <business outcome> on Intel edge" request: detect/count/track objects in camera feeds, spatial multi-camera analytics, video search & summarization, conversational Q&A / RAG over documents, multimodal embeddings, downloading/converting models, training a computer-vision model, or deploying a robot policy — when you do NOT already know which specific skill to run. DO NOT USE when the user already named a concrete skill (invoke that skill directly) or asks a pure code question with no deployable outcome.
Configure real-time Smart Community monitor alerts for a specified OpenClaw agent. Use when a user asks to bind, route, subscribe, or push one monitor's alerts to an agent session. Requires an agent ID and monitor ID, and always targets session agent:{agent-id}:{monitor-id}.
基于 SOC 职业分类
正在显示 SKILL.md
| name | lvc-test |
| description | Run the Live Video Captioning unit test suite (pytest in app/). Use when asked to test, verify, or check coverage of backend changes. |
Run from live-video-captioning/app/. Host Python 3.12 with deps already installed works; no venv needed on this box.
cd app
python3 -m pytest -q # full suite, ~3 s, 251 tests
python3 -m pytest -q tests/test_routes_runs.py # single file
--cov=backend --cov=main --cov-fail-under=80)./lvc-run-app.