[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-nvidia-vss-setup-behavior-analytics":3,"mdc-ixtl99-key":31,"related-org-nvidia-vss-setup-behavior-analytics":1144,"related-repo-nvidia-vss-setup-behavior-analytics":1303},{"slug":4,"name":4,"fn":5,"description":6,"org":7,"tags":11,"stars":20,"repoUrl":21,"updatedAt":22,"license":23,"forks":24,"topics":25,"repo":26,"sourceUrl":29,"mdContent":30},"vss-setup-behavior-analytics","deploy vss behavior analytics service","Use to deploy the vss-behavior-analytics service standalone (entrypoint, config-source, optional calibration). Not for the full warehouse deploy.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},"nvidia","NVIDIA","https:\u002F\u002Fpexgzepcugksgbtrxkhf.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Forg-logos\u002Fnvidia.png",[12,16,17],{"name":13,"slug":14,"type":15},"Deployment","deployment","tag",{"name":9,"slug":8,"type":15},{"name":18,"slug":19,"type":15},"Analytics","analytics",2473,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fskills","2026-07-14T05:26:23.970795","Apache-2.0",281,[],{"repoUrl":21,"stars":20,"forks":24,"topics":27,"description":28},[],"AI agent skills published by NVIDIA","https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fskills\u002Ftree\u002FHEAD\u002Fskills\u002Fvss-setup-behavior-analytics","---\nname: vss-setup-behavior-analytics\ndescription: Use to deploy the vss-behavior-analytics service standalone (entrypoint, config-source, optional calibration). Not for the full warehouse deploy.\nlicense: Apache-2.0\n\nmetadata:\n  author: \"NVIDIA Video Search and Summarization team\"\n  version: \"3.2.0\"\n  github-url: \"https:\u002F\u002Fgithub.com\u002FNVIDIA-AI-Blueprints\u002Fvideo-search-and-summarization\"\n  tags: \"nvidia blueprint operational deployment behavior-analytics\"\n---\n## Purpose\n\nDeploy the behavior-analytics service standalone with the user's chosen entrypoint, config, and calibration.\n\n## Instructions\n\nFollow the routing tables and step-by-step workflows below. Each section that ends in *workflow*, *quick start*, or *flow* is intended to be executed top-to-bottom. Detailed reference material lives in `references\u002F`.\n\n## Examples\n\nWorked end-to-end examples are kept under `evals\u002F` (each `*.json` manifest\ncontains a runnable scenario). Run a Tier-3 evaluation to replay them:\n\n```bash\nnv-base validate skills\u002Fvss-setup-behavior-analytics --agent-eval\n```\n\nA minimal standalone bring-up looks like:\n\n```bash\ncd $REPO\u002Fdeploy\u002Fdocker\nexport VSS_APPS_DIR=$(pwd)\ndocker compose -f services\u002Fanalytics\u002Fbehavior-analytics\u002Fcompose.yml up -d vss-behavior-analytics-base\n```\n\nFollow `references\u002Fdeploy-behavior-analytics-service.md` for the full\nworkflow (entrypoint pick, config source, dynamic updates).\n\n## Limitations\n\n- Requires the matching VSS profile \u002F microservice to be deployed and reachable from the caller.\n- NGC-hosted models and NIMs may be subject to rate-limits, GPU memory requirements, and license restrictions.\n- Concurrency, GPU memory, and storage limits depend on the host hardware and the profile's compose file.\n\n## Troubleshooting\n\n- **Error**: REST call returns connection refused. **Cause**: target microservice not running. **Solution**: probe `\u002Fdocs` or `\u002Fhealth`; redeploy via `vss-deploy-profile` or the matching `vss-deploy-*` skill.\n- **Error**: HTTP 401\u002F403 from NGC pulls. **Cause**: missing\u002Fexpired `NGC_CLI_API_KEY`. **Solution**: `docker login nvcr.io` and re-export the key before retrying.\n- **Error**: container OOM or model fails to load. **Cause**: insufficient GPU memory for the selected profile. **Solution**: switch to a smaller variant or free GPUs via `docker compose down`.\n\n# VSS Setup Behavior Analytics — Standalone\n\nDeploy **just** the `vss-behavior-analytics` container (the spatial-AI analytics pipeline from the upstream `behavior-analytics` repo), not as part of the full warehouse blueprint stack.\n\nThe full operational walkthrough — entrypoint table, config-source options, calibration types, dynamic-update wire contract, troubleshooting — is [`references\u002Fdeploy-behavior-analytics-service.md`](references\u002Fdeploy-behavior-analytics-service.md). This SKILL.md only handles routing and prerequisites.\n\n## When to use\n\n- \"Deploy behavior analytics\" \u002F \"run behavior-analytics standalone\"\n- \"I just want to run analytics, not the full stack\"\n- \"Change the entrypoint to fusion_search \u002F dev_example \u002F analytics 3D \u002F mv3dt\"\n- \"Use my own behavior-analytics config \u002F calibration JSON\"\n- \"Point behavior-analytics at the warehouse-3d (or mv3dt) config without spinning up the rest of the warehouse profile\"\n- \"Dynamic config \u002F dynamic calibration into a running behavior-analytics\"\n\n## Prerequisites\n\n1. **Repo checkout** with `$VSS_APPS_DIR` pointing at `\u003Crepo>\u002Fdeploy\u002Fdocker\u002F`. Required by the service compose's volume binds.\n2. **NGC credentials** — `$NGC_CLI_API_KEY` set so docker can pull the image. See [`references\u002Fngc-api-key-registry-login.md`](references\u002Fngc-api-key-registry-login.md).\n3. **Docker runtime** — Docker Engine **28.3.3** with Docker Compose plugin **v2.39.1+**. Verify with `docker --version` and `docker compose version`.\n4. **Optional broker** (Kafka \u002F Redis Streams \u002F MQTT). The container starts fine **without** one — the Kafka client retries a bounded number of times, then the app exits and `restart: always` cycles the container. Status will show `Restarting (N)` in `docker ps` until a broker is reachable. With a broker, dynamic config \u002F dynamic calibration over `mdx-notification` become available.\n5. **Optional config \u002F calibration files on disk** if the user is bringing their own.\n\nIf any required prerequisite fails, surface the gap before going further.\n\n## Workflow\n\nHand the user [`references\u002Fdeploy-behavior-analytics-service.md`](references\u002Fdeploy-behavior-analytics-service.md) and walk them through its steps in order:\n\n1. Pick an entrypoint (analytics 2D \u002F 3D \u002F mv3dt, dev_example, fusion_search).\n2. Choose a config — profile-shipped or custom.\n3. Choose a calibration — optional; profile-shipped or custom; otherwise the app waits for a dynamic-calibration notification.\n4. Decide whether a broker is reachable; if yes, point them at the dynamic-update flows.\n\nThe compose-file edits, YAML diffs, deploy + verify commands, and troubleshooting table all live in that reference — don't duplicate them here.\n\n## Dynamic updates (runtime, no restart)\n\nOnce the container is up **and a broker is reachable**, two runtime-update flows are available — neither requires redeploying:\n\n### Dynamic config\n\nPublish an `upsert` (per-key patch) or `upsert-all` (full snapshot) message to the `mdx-notification` topic with Kafka key `behavior-analytics-config` and headers:\n\n- `event.type`: `upsert` | `upsert-all` | `request-config` | `ack`\n- `reference-id`: `video-analytics-api-\u003Cuuid>` (web-api originated), `behavior-analytics-\u003Cuuid>` (bootstrap reply), or the source-type literal (`kafka` \u002F `redis` \u002F `mqtt`) for direct-publisher upserts.\n\nBody: `{\"status\": ..., \"config\": \u003Cpatch>, \"error\": ...}`.\n\nThe listener validates each message at the envelope layer (rejects unknown keys, missing config, malformed status\u002Ferror) and at the per-payload layer (rejects forbidden sections, bad item shapes). Successful upserts are persisted to disk, applied to every worker, and ACK'd back over the topic.\n\nFull wire contract + ack semantics: [`references\u002Fdynamic-config.md`](references\u002Fdynamic-config.md).\n\n### Dynamic calibration\n\nPublish to the same topic with Kafka key `calibration` and headers:\n\n- `event.type`: `upsert-all` (full snapshot) | `upsert` (per-sensor merge) | `delete` (per-sensor removal)\n- `timestamp`: ISO-8601 UTC (`YYYY-MM-DDTHH:MM:SS.fffZ`).\n\nBody: JSON sensor list (and ROIs \u002F tripwires \u002F homographies for `upsert-all`).\n\nThe listener validates against the vendored AJV schema before persisting. Schema violations log a `calibration schema violation` warning and are dropped — the previously-good calibration stays loaded.\n\nFull wire contract + per-action validation policy: [`references\u002Fdynamic-calibration.md`](references\u002Fdynamic-calibration.md).\n\nBoth flows live entirely on the broker — the producer can be `video-analytics-api`, your own script, or any Kafka client that mirrors the wire shape. They're the recommended way to change configuration after the container is running, so the operator doesn't have to redeploy.\n\n## Routing rules\n\n- If the user wants \"the full stack\" (UI \u002F agent \u002F perception): hand off to [`vss-deploy-profile`](..\u002Fvss-deploy-profile\u002FSKILL.md) with profile `warehouse` (or `alerts`). Don't run this skill in parallel.\n- If the user wants to publish a runtime config \u002F calibration update to an already-running container: walk the [Dynamic updates](#dynamic-updates-runtime-no-restart) section. Both flows need a reachable broker.\n- If the user describes a behavior-analytics behavior change they want to validate (new incident type, new ROI rule, new sensor): point them at [`references\u002Fconfiguration.md`](references\u002Fconfiguration.md), [`references\u002Fdynamic-config.md`](references\u002Fdynamic-config.md), or [`references\u002Fdynamic-calibration.md`](references\u002Fdynamic-calibration.md) before editing the JSON.\n\nbump:1\n",{"data":32,"body":38},{"name":4,"description":6,"license":23,"metadata":33},{"author":34,"version":35,"github-url":36,"tags":37},"NVIDIA Video Search and Summarization team","3.2.0","https:\u002F\u002Fgithub.com\u002FNVIDIA-AI-Blueprints\u002Fvideo-search-and-summarization","nvidia blueprint operational deployment behavior-analytics",{"type":39,"children":40},"root",[41,50,56,62,98,104,125,164,169,266,279,285,305,311,436,443,471,487,493,526,532,688,693,698,713,736,741,747,759,766,802,890,902,907,922,928,940,993,1004,1017,1032,1045,1051,1133,1138],{"type":42,"tag":43,"props":44,"children":46},"element","h2",{"id":45},"purpose",[47],{"type":48,"value":49},"text","Purpose",{"type":42,"tag":51,"props":52,"children":53},"p",{},[54],{"type":48,"value":55},"Deploy the behavior-analytics service standalone with the user's chosen entrypoint, config, and calibration.",{"type":42,"tag":43,"props":57,"children":59},{"id":58},"instructions",[60],{"type":48,"value":61},"Instructions",{"type":42,"tag":51,"props":63,"children":64},{},[65,67,73,75,80,82,87,89,96],{"type":48,"value":66},"Follow the routing tables and step-by-step workflows below. Each section that ends in ",{"type":42,"tag":68,"props":69,"children":70},"em",{},[71],{"type":48,"value":72},"workflow",{"type":48,"value":74},", ",{"type":42,"tag":68,"props":76,"children":77},{},[78],{"type":48,"value":79},"quick start",{"type":48,"value":81},", or ",{"type":42,"tag":68,"props":83,"children":84},{},[85],{"type":48,"value":86},"flow",{"type":48,"value":88}," is intended to be executed top-to-bottom. Detailed reference material lives in ",{"type":42,"tag":90,"props":91,"children":93},"code",{"className":92},[],[94],{"type":48,"value":95},"references\u002F",{"type":48,"value":97},".",{"type":42,"tag":43,"props":99,"children":101},{"id":100},"examples",[102],{"type":48,"value":103},"Examples",{"type":42,"tag":51,"props":105,"children":106},{},[107,109,115,117,123],{"type":48,"value":108},"Worked end-to-end examples are kept under ",{"type":42,"tag":90,"props":110,"children":112},{"className":111},[],[113],{"type":48,"value":114},"evals\u002F",{"type":48,"value":116}," (each ",{"type":42,"tag":90,"props":118,"children":120},{"className":119},[],[121],{"type":48,"value":122},"*.json",{"type":48,"value":124}," manifest\ncontains a runnable scenario). Run a Tier-3 evaluation to replay them:",{"type":42,"tag":126,"props":127,"children":132},"pre",{"className":128,"code":129,"language":130,"meta":131,"style":131},"language-bash shiki shiki-themes material-theme-lighter material-theme material-theme-palenight","nv-base validate skills\u002Fvss-setup-behavior-analytics --agent-eval\n","bash","",[133],{"type":42,"tag":90,"props":134,"children":135},{"__ignoreMap":131},[136],{"type":42,"tag":137,"props":138,"children":141},"span",{"class":139,"line":140},"line",1,[142,148,154,159],{"type":42,"tag":137,"props":143,"children":145},{"style":144},"--shiki-light:#E2931D;--shiki-default:#FFCB6B;--shiki-dark:#FFCB6B",[146],{"type":48,"value":147},"nv-base",{"type":42,"tag":137,"props":149,"children":151},{"style":150},"--shiki-light:#91B859;--shiki-default:#C3E88D;--shiki-dark:#C3E88D",[152],{"type":48,"value":153}," validate",{"type":42,"tag":137,"props":155,"children":156},{"style":150},[157],{"type":48,"value":158}," skills\u002Fvss-setup-behavior-analytics",{"type":42,"tag":137,"props":160,"children":161},{"style":150},[162],{"type":48,"value":163}," --agent-eval\n",{"type":42,"tag":51,"props":165,"children":166},{},[167],{"type":48,"value":168},"A minimal standalone bring-up looks like:",{"type":42,"tag":126,"props":170,"children":172},{"className":128,"code":171,"language":130,"meta":131,"style":131},"cd $REPO\u002Fdeploy\u002Fdocker\nexport VSS_APPS_DIR=$(pwd)\ndocker compose -f services\u002Fanalytics\u002Fbehavior-analytics\u002Fcompose.yml up -d vss-behavior-analytics-base\n",[173],{"type":42,"tag":90,"props":174,"children":175},{"__ignoreMap":131},[176,196,227],{"type":42,"tag":137,"props":177,"children":178},{"class":139,"line":140},[179,185,191],{"type":42,"tag":137,"props":180,"children":182},{"style":181},"--shiki-light:#6182B8;--shiki-default:#82AAFF;--shiki-dark:#82AAFF",[183],{"type":48,"value":184},"cd",{"type":42,"tag":137,"props":186,"children":188},{"style":187},"--shiki-light:#90A4AE;--shiki-default:#EEFFFF;--shiki-dark:#BABED8",[189],{"type":48,"value":190}," $REPO",{"type":42,"tag":137,"props":192,"children":193},{"style":150},[194],{"type":48,"value":195},"\u002Fdeploy\u002Fdocker\n",{"type":42,"tag":137,"props":197,"children":199},{"class":139,"line":198},2,[200,206,211,217,222],{"type":42,"tag":137,"props":201,"children":203},{"style":202},"--shiki-light:#9C3EDA;--shiki-default:#C792EA;--shiki-dark:#C792EA",[204],{"type":48,"value":205},"export",{"type":42,"tag":137,"props":207,"children":208},{"style":187},[209],{"type":48,"value":210}," VSS_APPS_DIR",{"type":42,"tag":137,"props":212,"children":214},{"style":213},"--shiki-light:#39ADB5;--shiki-default:#89DDFF;--shiki-dark:#89DDFF",[215],{"type":48,"value":216},"=$(",{"type":42,"tag":137,"props":218,"children":219},{"style":181},[220],{"type":48,"value":221},"pwd",{"type":42,"tag":137,"props":223,"children":224},{"style":213},[225],{"type":48,"value":226},")\n",{"type":42,"tag":137,"props":228,"children":230},{"class":139,"line":229},3,[231,236,241,246,251,256,261],{"type":42,"tag":137,"props":232,"children":233},{"style":144},[234],{"type":48,"value":235},"docker",{"type":42,"tag":137,"props":237,"children":238},{"style":150},[239],{"type":48,"value":240}," compose",{"type":42,"tag":137,"props":242,"children":243},{"style":150},[244],{"type":48,"value":245}," -f",{"type":42,"tag":137,"props":247,"children":248},{"style":150},[249],{"type":48,"value":250}," services\u002Fanalytics\u002Fbehavior-analytics\u002Fcompose.yml",{"type":42,"tag":137,"props":252,"children":253},{"style":150},[254],{"type":48,"value":255}," up",{"type":42,"tag":137,"props":257,"children":258},{"style":150},[259],{"type":48,"value":260}," -d",{"type":42,"tag":137,"props":262,"children":263},{"style":150},[264],{"type":48,"value":265}," vss-behavior-analytics-base\n",{"type":42,"tag":51,"props":267,"children":268},{},[269,271,277],{"type":48,"value":270},"Follow ",{"type":42,"tag":90,"props":272,"children":274},{"className":273},[],[275],{"type":48,"value":276},"references\u002Fdeploy-behavior-analytics-service.md",{"type":48,"value":278}," for the full\nworkflow (entrypoint pick, config source, dynamic updates).",{"type":42,"tag":43,"props":280,"children":282},{"id":281},"limitations",[283],{"type":48,"value":284},"Limitations",{"type":42,"tag":286,"props":287,"children":288},"ul",{},[289,295,300],{"type":42,"tag":290,"props":291,"children":292},"li",{},[293],{"type":48,"value":294},"Requires the matching VSS profile \u002F microservice to be deployed and reachable from the caller.",{"type":42,"tag":290,"props":296,"children":297},{},[298],{"type":48,"value":299},"NGC-hosted models and NIMs may be subject to rate-limits, GPU memory requirements, and license restrictions.",{"type":42,"tag":290,"props":301,"children":302},{},[303],{"type":48,"value":304},"Concurrency, GPU memory, and storage limits depend on the host hardware and the profile's compose file.",{"type":42,"tag":43,"props":306,"children":308},{"id":307},"troubleshooting",[309],{"type":48,"value":310},"Troubleshooting",{"type":42,"tag":286,"props":312,"children":313},{},[314,371,408],{"type":42,"tag":290,"props":315,"children":316},{},[317,323,325,330,332,337,339,345,347,353,355,361,363,369],{"type":42,"tag":318,"props":319,"children":320},"strong",{},[321],{"type":48,"value":322},"Error",{"type":48,"value":324},": REST call returns connection refused. ",{"type":42,"tag":318,"props":326,"children":327},{},[328],{"type":48,"value":329},"Cause",{"type":48,"value":331},": target microservice not running. ",{"type":42,"tag":318,"props":333,"children":334},{},[335],{"type":48,"value":336},"Solution",{"type":48,"value":338},": probe ",{"type":42,"tag":90,"props":340,"children":342},{"className":341},[],[343],{"type":48,"value":344},"\u002Fdocs",{"type":48,"value":346}," or ",{"type":42,"tag":90,"props":348,"children":350},{"className":349},[],[351],{"type":48,"value":352},"\u002Fhealth",{"type":48,"value":354},"; redeploy via ",{"type":42,"tag":90,"props":356,"children":358},{"className":357},[],[359],{"type":48,"value":360},"vss-deploy-profile",{"type":48,"value":362}," or the matching ",{"type":42,"tag":90,"props":364,"children":366},{"className":365},[],[367],{"type":48,"value":368},"vss-deploy-*",{"type":48,"value":370}," skill.",{"type":42,"tag":290,"props":372,"children":373},{},[374,378,380,384,386,392,394,398,400,406],{"type":42,"tag":318,"props":375,"children":376},{},[377],{"type":48,"value":322},{"type":48,"value":379},": HTTP 401\u002F403 from NGC pulls. ",{"type":42,"tag":318,"props":381,"children":382},{},[383],{"type":48,"value":329},{"type":48,"value":385},": missing\u002Fexpired ",{"type":42,"tag":90,"props":387,"children":389},{"className":388},[],[390],{"type":48,"value":391},"NGC_CLI_API_KEY",{"type":48,"value":393},". ",{"type":42,"tag":318,"props":395,"children":396},{},[397],{"type":48,"value":336},{"type":48,"value":399},": ",{"type":42,"tag":90,"props":401,"children":403},{"className":402},[],[404],{"type":48,"value":405},"docker login nvcr.io",{"type":48,"value":407}," and re-export the key before retrying.",{"type":42,"tag":290,"props":409,"children":410},{},[411,415,417,421,423,427,429,435],{"type":42,"tag":318,"props":412,"children":413},{},[414],{"type":48,"value":322},{"type":48,"value":416},": container OOM or model fails to load. ",{"type":42,"tag":318,"props":418,"children":419},{},[420],{"type":48,"value":329},{"type":48,"value":422},": insufficient GPU memory for the selected profile. ",{"type":42,"tag":318,"props":424,"children":425},{},[426],{"type":48,"value":336},{"type":48,"value":428},": switch to a smaller variant or free GPUs via ",{"type":42,"tag":90,"props":430,"children":432},{"className":431},[],[433],{"type":48,"value":434},"docker compose down",{"type":48,"value":97},{"type":42,"tag":437,"props":438,"children":440},"h1",{"id":439},"vss-setup-behavior-analytics-standalone",[441],{"type":48,"value":442},"VSS Setup Behavior Analytics — Standalone",{"type":42,"tag":51,"props":444,"children":445},{},[446,448,453,455,461,463,469],{"type":48,"value":447},"Deploy ",{"type":42,"tag":318,"props":449,"children":450},{},[451],{"type":48,"value":452},"just",{"type":48,"value":454}," the ",{"type":42,"tag":90,"props":456,"children":458},{"className":457},[],[459],{"type":48,"value":460},"vss-behavior-analytics",{"type":48,"value":462}," container (the spatial-AI analytics pipeline from the upstream ",{"type":42,"tag":90,"props":464,"children":466},{"className":465},[],[467],{"type":48,"value":468},"behavior-analytics",{"type":48,"value":470}," repo), not as part of the full warehouse blueprint stack.",{"type":42,"tag":51,"props":472,"children":473},{},[474,476,485],{"type":48,"value":475},"The full operational walkthrough — entrypoint table, config-source options, calibration types, dynamic-update wire contract, troubleshooting — is ",{"type":42,"tag":477,"props":478,"children":479},"a",{"href":276},[480],{"type":42,"tag":90,"props":481,"children":483},{"className":482},[],[484],{"type":48,"value":276},{"type":48,"value":486},". This SKILL.md only handles routing and prerequisites.",{"type":42,"tag":43,"props":488,"children":490},{"id":489},"when-to-use",[491],{"type":48,"value":492},"When to use",{"type":42,"tag":286,"props":494,"children":495},{},[496,501,506,511,516,521],{"type":42,"tag":290,"props":497,"children":498},{},[499],{"type":48,"value":500},"\"Deploy behavior analytics\" \u002F \"run behavior-analytics standalone\"",{"type":42,"tag":290,"props":502,"children":503},{},[504],{"type":48,"value":505},"\"I just want to run analytics, not the full stack\"",{"type":42,"tag":290,"props":507,"children":508},{},[509],{"type":48,"value":510},"\"Change the entrypoint to fusion_search \u002F dev_example \u002F analytics 3D \u002F mv3dt\"",{"type":42,"tag":290,"props":512,"children":513},{},[514],{"type":48,"value":515},"\"Use my own behavior-analytics config \u002F calibration JSON\"",{"type":42,"tag":290,"props":517,"children":518},{},[519],{"type":48,"value":520},"\"Point behavior-analytics at the warehouse-3d (or mv3dt) config without spinning up the rest of the warehouse profile\"",{"type":42,"tag":290,"props":522,"children":523},{},[524],{"type":48,"value":525},"\"Dynamic config \u002F dynamic calibration into a running behavior-analytics\"",{"type":42,"tag":43,"props":527,"children":529},{"id":528},"prerequisites",[530],{"type":48,"value":531},"Prerequisites",{"type":42,"tag":533,"props":534,"children":535},"ol",{},[536,562,590,629,678],{"type":42,"tag":290,"props":537,"children":538},{},[539,544,546,552,554,560],{"type":42,"tag":318,"props":540,"children":541},{},[542],{"type":48,"value":543},"Repo checkout",{"type":48,"value":545}," with ",{"type":42,"tag":90,"props":547,"children":549},{"className":548},[],[550],{"type":48,"value":551},"$VSS_APPS_DIR",{"type":48,"value":553}," pointing at ",{"type":42,"tag":90,"props":555,"children":557},{"className":556},[],[558],{"type":48,"value":559},"\u003Crepo>\u002Fdeploy\u002Fdocker\u002F",{"type":48,"value":561},". Required by the service compose's volume binds.",{"type":42,"tag":290,"props":563,"children":564},{},[565,570,572,578,580,589],{"type":42,"tag":318,"props":566,"children":567},{},[568],{"type":48,"value":569},"NGC credentials",{"type":48,"value":571}," — ",{"type":42,"tag":90,"props":573,"children":575},{"className":574},[],[576],{"type":48,"value":577},"$NGC_CLI_API_KEY",{"type":48,"value":579}," set so docker can pull the image. See ",{"type":42,"tag":477,"props":581,"children":583},{"href":582},"references\u002Fngc-api-key-registry-login.md",[584],{"type":42,"tag":90,"props":585,"children":587},{"className":586},[],[588],{"type":48,"value":582},{"type":48,"value":97},{"type":42,"tag":290,"props":591,"children":592},{},[593,598,600,605,607,612,614,620,622,628],{"type":42,"tag":318,"props":594,"children":595},{},[596],{"type":48,"value":597},"Docker runtime",{"type":48,"value":599}," — Docker Engine ",{"type":42,"tag":318,"props":601,"children":602},{},[603],{"type":48,"value":604},"28.3.3",{"type":48,"value":606}," with Docker Compose plugin ",{"type":42,"tag":318,"props":608,"children":609},{},[610],{"type":48,"value":611},"v2.39.1+",{"type":48,"value":613},". Verify with ",{"type":42,"tag":90,"props":615,"children":617},{"className":616},[],[618],{"type":48,"value":619},"docker --version",{"type":48,"value":621}," and ",{"type":42,"tag":90,"props":623,"children":625},{"className":624},[],[626],{"type":48,"value":627},"docker compose version",{"type":48,"value":97},{"type":42,"tag":290,"props":630,"children":631},{},[632,637,639,644,646,652,654,660,662,668,670,676],{"type":42,"tag":318,"props":633,"children":634},{},[635],{"type":48,"value":636},"Optional broker",{"type":48,"value":638}," (Kafka \u002F Redis Streams \u002F MQTT). The container starts fine ",{"type":42,"tag":318,"props":640,"children":641},{},[642],{"type":48,"value":643},"without",{"type":48,"value":645}," one — the Kafka client retries a bounded number of times, then the app exits and ",{"type":42,"tag":90,"props":647,"children":649},{"className":648},[],[650],{"type":48,"value":651},"restart: always",{"type":48,"value":653}," cycles the container. Status will show ",{"type":42,"tag":90,"props":655,"children":657},{"className":656},[],[658],{"type":48,"value":659},"Restarting (N)",{"type":48,"value":661}," in ",{"type":42,"tag":90,"props":663,"children":665},{"className":664},[],[666],{"type":48,"value":667},"docker ps",{"type":48,"value":669}," until a broker is reachable. With a broker, dynamic config \u002F dynamic calibration over ",{"type":42,"tag":90,"props":671,"children":673},{"className":672},[],[674],{"type":48,"value":675},"mdx-notification",{"type":48,"value":677}," become available.",{"type":42,"tag":290,"props":679,"children":680},{},[681,686],{"type":42,"tag":318,"props":682,"children":683},{},[684],{"type":48,"value":685},"Optional config \u002F calibration files on disk",{"type":48,"value":687}," if the user is bringing their own.",{"type":42,"tag":51,"props":689,"children":690},{},[691],{"type":48,"value":692},"If any required prerequisite fails, surface the gap before going further.",{"type":42,"tag":43,"props":694,"children":695},{"id":72},[696],{"type":48,"value":697},"Workflow",{"type":42,"tag":51,"props":699,"children":700},{},[701,703,711],{"type":48,"value":702},"Hand the user ",{"type":42,"tag":477,"props":704,"children":705},{"href":276},[706],{"type":42,"tag":90,"props":707,"children":709},{"className":708},[],[710],{"type":48,"value":276},{"type":48,"value":712}," and walk them through its steps in order:",{"type":42,"tag":533,"props":714,"children":715},{},[716,721,726,731],{"type":42,"tag":290,"props":717,"children":718},{},[719],{"type":48,"value":720},"Pick an entrypoint (analytics 2D \u002F 3D \u002F mv3dt, dev_example, fusion_search).",{"type":42,"tag":290,"props":722,"children":723},{},[724],{"type":48,"value":725},"Choose a config — profile-shipped or custom.",{"type":42,"tag":290,"props":727,"children":728},{},[729],{"type":48,"value":730},"Choose a calibration — optional; profile-shipped or custom; otherwise the app waits for a dynamic-calibration notification.",{"type":42,"tag":290,"props":732,"children":733},{},[734],{"type":48,"value":735},"Decide whether a broker is reachable; if yes, point them at the dynamic-update flows.",{"type":42,"tag":51,"props":737,"children":738},{},[739],{"type":48,"value":740},"The compose-file edits, YAML diffs, deploy + verify commands, and troubleshooting table all live in that reference — don't duplicate them here.",{"type":42,"tag":43,"props":742,"children":744},{"id":743},"dynamic-updates-runtime-no-restart",[745],{"type":48,"value":746},"Dynamic updates (runtime, no restart)",{"type":42,"tag":51,"props":748,"children":749},{},[750,752,757],{"type":48,"value":751},"Once the container is up ",{"type":42,"tag":318,"props":753,"children":754},{},[755],{"type":48,"value":756},"and a broker is reachable",{"type":48,"value":758},", two runtime-update flows are available — neither requires redeploying:",{"type":42,"tag":760,"props":761,"children":763},"h3",{"id":762},"dynamic-config",[764],{"type":48,"value":765},"Dynamic config",{"type":42,"tag":51,"props":767,"children":768},{},[769,771,777,779,785,787,792,794,800],{"type":48,"value":770},"Publish an ",{"type":42,"tag":90,"props":772,"children":774},{"className":773},[],[775],{"type":48,"value":776},"upsert",{"type":48,"value":778}," (per-key patch) or ",{"type":42,"tag":90,"props":780,"children":782},{"className":781},[],[783],{"type":48,"value":784},"upsert-all",{"type":48,"value":786}," (full snapshot) message to the ",{"type":42,"tag":90,"props":788,"children":790},{"className":789},[],[791],{"type":48,"value":675},{"type":48,"value":793}," topic with Kafka key ",{"type":42,"tag":90,"props":795,"children":797},{"className":796},[],[798],{"type":48,"value":799},"behavior-analytics-config",{"type":48,"value":801}," and headers:",{"type":42,"tag":286,"props":803,"children":804},{},[805,841],{"type":42,"tag":290,"props":806,"children":807},{},[808,814,815,820,822,827,828,834,835],{"type":42,"tag":90,"props":809,"children":811},{"className":810},[],[812],{"type":48,"value":813},"event.type",{"type":48,"value":399},{"type":42,"tag":90,"props":816,"children":818},{"className":817},[],[819],{"type":48,"value":776},{"type":48,"value":821}," | ",{"type":42,"tag":90,"props":823,"children":825},{"className":824},[],[826],{"type":48,"value":784},{"type":48,"value":821},{"type":42,"tag":90,"props":829,"children":831},{"className":830},[],[832],{"type":48,"value":833},"request-config",{"type":48,"value":821},{"type":42,"tag":90,"props":836,"children":838},{"className":837},[],[839],{"type":48,"value":840},"ack",{"type":42,"tag":290,"props":842,"children":843},{},[844,850,851,857,859,865,867,873,875,881,882,888],{"type":42,"tag":90,"props":845,"children":847},{"className":846},[],[848],{"type":48,"value":849},"reference-id",{"type":48,"value":399},{"type":42,"tag":90,"props":852,"children":854},{"className":853},[],[855],{"type":48,"value":856},"video-analytics-api-\u003Cuuid>",{"type":48,"value":858}," (web-api originated), ",{"type":42,"tag":90,"props":860,"children":862},{"className":861},[],[863],{"type":48,"value":864},"behavior-analytics-\u003Cuuid>",{"type":48,"value":866}," (bootstrap reply), or the source-type literal (",{"type":42,"tag":90,"props":868,"children":870},{"className":869},[],[871],{"type":48,"value":872},"kafka",{"type":48,"value":874}," \u002F ",{"type":42,"tag":90,"props":876,"children":878},{"className":877},[],[879],{"type":48,"value":880},"redis",{"type":48,"value":874},{"type":42,"tag":90,"props":883,"children":885},{"className":884},[],[886],{"type":48,"value":887},"mqtt",{"type":48,"value":889},") for direct-publisher upserts.",{"type":42,"tag":51,"props":891,"children":892},{},[893,895,901],{"type":48,"value":894},"Body: ",{"type":42,"tag":90,"props":896,"children":898},{"className":897},[],[899],{"type":48,"value":900},"{\"status\": ..., \"config\": \u003Cpatch>, \"error\": ...}",{"type":48,"value":97},{"type":42,"tag":51,"props":903,"children":904},{},[905],{"type":48,"value":906},"The listener validates each message at the envelope layer (rejects unknown keys, missing config, malformed status\u002Ferror) and at the per-payload layer (rejects forbidden sections, bad item shapes). Successful upserts are persisted to disk, applied to every worker, and ACK'd back over the topic.",{"type":42,"tag":51,"props":908,"children":909},{},[910,912,921],{"type":48,"value":911},"Full wire contract + ack semantics: ",{"type":42,"tag":477,"props":913,"children":915},{"href":914},"references\u002Fdynamic-config.md",[916],{"type":42,"tag":90,"props":917,"children":919},{"className":918},[],[920],{"type":48,"value":914},{"type":48,"value":97},{"type":42,"tag":760,"props":923,"children":925},{"id":924},"dynamic-calibration",[926],{"type":48,"value":927},"Dynamic calibration",{"type":42,"tag":51,"props":929,"children":930},{},[931,933,939],{"type":48,"value":932},"Publish to the same topic with Kafka key ",{"type":42,"tag":90,"props":934,"children":936},{"className":935},[],[937],{"type":48,"value":938},"calibration",{"type":48,"value":801},{"type":42,"tag":286,"props":941,"children":942},{},[943,974],{"type":42,"tag":290,"props":944,"children":945},{},[946,951,952,957,959,964,966,972],{"type":42,"tag":90,"props":947,"children":949},{"className":948},[],[950],{"type":48,"value":813},{"type":48,"value":399},{"type":42,"tag":90,"props":953,"children":955},{"className":954},[],[956],{"type":48,"value":784},{"type":48,"value":958}," (full snapshot) | ",{"type":42,"tag":90,"props":960,"children":962},{"className":961},[],[963],{"type":48,"value":776},{"type":48,"value":965}," (per-sensor merge) | ",{"type":42,"tag":90,"props":967,"children":969},{"className":968},[],[970],{"type":48,"value":971},"delete",{"type":48,"value":973}," (per-sensor removal)",{"type":42,"tag":290,"props":975,"children":976},{},[977,983,985,991],{"type":42,"tag":90,"props":978,"children":980},{"className":979},[],[981],{"type":48,"value":982},"timestamp",{"type":48,"value":984},": ISO-8601 UTC (",{"type":42,"tag":90,"props":986,"children":988},{"className":987},[],[989],{"type":48,"value":990},"YYYY-MM-DDTHH:MM:SS.fffZ",{"type":48,"value":992},").",{"type":42,"tag":51,"props":994,"children":995},{},[996,998,1003],{"type":48,"value":997},"Body: JSON sensor list (and ROIs \u002F tripwires \u002F homographies for ",{"type":42,"tag":90,"props":999,"children":1001},{"className":1000},[],[1002],{"type":48,"value":784},{"type":48,"value":992},{"type":42,"tag":51,"props":1005,"children":1006},{},[1007,1009,1015],{"type":48,"value":1008},"The listener validates against the vendored AJV schema before persisting. Schema violations log a ",{"type":42,"tag":90,"props":1010,"children":1012},{"className":1011},[],[1013],{"type":48,"value":1014},"calibration schema violation",{"type":48,"value":1016}," warning and are dropped — the previously-good calibration stays loaded.",{"type":42,"tag":51,"props":1018,"children":1019},{},[1020,1022,1031],{"type":48,"value":1021},"Full wire contract + per-action validation policy: ",{"type":42,"tag":477,"props":1023,"children":1025},{"href":1024},"references\u002Fdynamic-calibration.md",[1026],{"type":42,"tag":90,"props":1027,"children":1029},{"className":1028},[],[1030],{"type":48,"value":1024},{"type":48,"value":97},{"type":42,"tag":51,"props":1033,"children":1034},{},[1035,1037,1043],{"type":48,"value":1036},"Both flows live entirely on the broker — the producer can be ",{"type":42,"tag":90,"props":1038,"children":1040},{"className":1039},[],[1041],{"type":48,"value":1042},"video-analytics-api",{"type":48,"value":1044},", your own script, or any Kafka client that mirrors the wire shape. They're the recommended way to change configuration after the container is running, so the operator doesn't have to redeploy.",{"type":42,"tag":43,"props":1046,"children":1048},{"id":1047},"routing-rules",[1049],{"type":48,"value":1050},"Routing rules",{"type":42,"tag":286,"props":1052,"children":1053},{},[1054,1086,1099],{"type":42,"tag":290,"props":1055,"children":1056},{},[1057,1059,1068,1070,1076,1078,1084],{"type":48,"value":1058},"If the user wants \"the full stack\" (UI \u002F agent \u002F perception): hand off to ",{"type":42,"tag":477,"props":1060,"children":1062},{"href":1061},"..\u002Fvss-deploy-profile\u002FSKILL.md",[1063],{"type":42,"tag":90,"props":1064,"children":1066},{"className":1065},[],[1067],{"type":48,"value":360},{"type":48,"value":1069}," with profile ",{"type":42,"tag":90,"props":1071,"children":1073},{"className":1072},[],[1074],{"type":48,"value":1075},"warehouse",{"type":48,"value":1077}," (or ",{"type":42,"tag":90,"props":1079,"children":1081},{"className":1080},[],[1082],{"type":48,"value":1083},"alerts",{"type":48,"value":1085},"). Don't run this skill in parallel.",{"type":42,"tag":290,"props":1087,"children":1088},{},[1089,1091,1097],{"type":48,"value":1090},"If the user wants to publish a runtime config \u002F calibration update to an already-running container: walk the ",{"type":42,"tag":477,"props":1092,"children":1094},{"href":1093},"#dynamic-updates-runtime-no-restart",[1095],{"type":48,"value":1096},"Dynamic updates",{"type":48,"value":1098}," section. Both flows need a reachable broker.",{"type":42,"tag":290,"props":1100,"children":1101},{},[1102,1104,1113,1114,1122,1123,1131],{"type":48,"value":1103},"If the user describes a behavior-analytics behavior change they want to validate (new incident type, new ROI rule, new sensor): point them at ",{"type":42,"tag":477,"props":1105,"children":1107},{"href":1106},"references\u002Fconfiguration.md",[1108],{"type":42,"tag":90,"props":1109,"children":1111},{"className":1110},[],[1112],{"type":48,"value":1106},{"type":48,"value":74},{"type":42,"tag":477,"props":1115,"children":1116},{"href":914},[1117],{"type":42,"tag":90,"props":1118,"children":1120},{"className":1119},[],[1121],{"type":48,"value":914},{"type":48,"value":81},{"type":42,"tag":477,"props":1124,"children":1125},{"href":1024},[1126],{"type":42,"tag":90,"props":1127,"children":1129},{"className":1128},[],[1130],{"type":48,"value":1024},{"type":48,"value":1132}," before editing the JSON.",{"type":42,"tag":51,"props":1134,"children":1135},{},[1136],{"type":48,"value":1137},"bump:1",{"type":42,"tag":1139,"props":1140,"children":1141},"style",{},[1142],{"type":48,"value":1143},"html .light .shiki span {color: var(--shiki-light);background: var(--shiki-light-bg);font-style: var(--shiki-light-font-style);font-weight: var(--shiki-light-font-weight);text-decoration: var(--shiki-light-text-decoration);}html.light .shiki span {color: var(--shiki-light);background: var(--shiki-light-bg);font-style: var(--shiki-light-font-style);font-weight: var(--shiki-light-font-weight);text-decoration: var(--shiki-light-text-decoration);}html .default .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .dark .shiki span {color: var(--shiki-dark);background: var(--shiki-dark-bg);font-style: var(--shiki-dark-font-style);font-weight: var(--shiki-dark-font-weight);text-decoration: var(--shiki-dark-text-decoration);}html.dark .shiki span {color: var(--shiki-dark);background: var(--shiki-dark-bg);font-style: var(--shiki-dark-font-style);font-weight: var(--shiki-dark-font-weight);text-decoration: var(--shiki-dark-text-decoration);}",{"items":1145,"total":1302},[1146,1164,1180,1191,1203,1217,1230,1244,1257,1268,1282,1291],{"slug":1147,"name":1147,"fn":1148,"description":1149,"org":1150,"tags":1151,"stars":1161,"repoUrl":1162,"updatedAt":1163},"nemoclaw-user-guide","retrieve NemoClaw documentation and configuration","Guides human users' AI agents to the NemoClaw docs MCP server and canonical Fern documentation in Markdown form. Use when users ask how to install, configure, operate, troubleshoot, secure, or learn NemoClaw with an AI coding assistant. Trigger keywords - nemoclaw docs, use nemoclaw with ai agent, nemoclaw mcp docs, nemoclaw install help, nemoclaw quickstart, nemoclaw markdown docs, llms.txt, agent skills.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1152,1155,1158],{"name":1153,"slug":1154,"type":15},"Documentation","documentation",{"name":1156,"slug":1157,"type":15},"MCP","mcp",{"name":1159,"slug":1160,"type":15},"Search","search",21777,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FNemoClaw","2026-07-20T06:00:01.461044",{"slug":1165,"name":1165,"fn":1166,"description":1167,"org":1168,"tags":1169,"stars":1177,"repoUrl":1178,"updatedAt":1179},"mcore-build-and-dependency","manage Megatron-LM development environments","Container-based dev environment setup and dependency management for Megatron-LM. Covers acquiring and launching the CI container, uv package management, and updating uv.lock.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1170,1173,1174],{"name":1171,"slug":1172,"type":15},"Containers","containers",{"name":13,"slug":14,"type":15},{"name":1175,"slug":1176,"type":15},"Python","python",17049,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FMegatron-LM","2026-07-27T06:06:11.249662",{"slug":1181,"name":1181,"fn":1182,"description":1183,"org":1184,"tags":1185,"stars":1177,"repoUrl":1178,"updatedAt":1190},"mcore-bump-base-image","update NVIDIA PyTorch base images","Bump the NVIDIA PyTorch base image (`nvcr.io\u002Fnvidia\u002Fpytorch:YY.MM-py3`) used by Megatron-LM CI. Covers the two pin sites (GitHub CI in `docker\u002F.ngc_version.dev` and GitLab CI in `.gitlab\u002Fstages\u002F01.build.yml`), the post-bump CI loop (re-run functional tests, refresh golden values, mark broken tests), and the gotchas that bit PRs",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1186,1189],{"name":1187,"slug":1188,"type":15},"CI\u002FCD","ci-cd",{"name":13,"slug":14,"type":15},"2026-07-14T05:25:59.97109",{"slug":1192,"name":1192,"fn":1193,"description":1194,"org":1195,"tags":1196,"stars":1177,"repoUrl":1178,"updatedAt":1202},"mcore-cicd","manage CI\u002FCD pipelines for Megatron-LM","CI\u002FCD reference for Megatron-LM. Covers CI pipeline structure, PR scope labels, triggering internal GitLab CI (which force-pushes the current branch to a pull-request\u002FBRANCH ref — always dry-run and verify the destination first; never run against shared or protected branches), and CI failure investigation.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1197,1198,1199],{"name":1187,"slug":1188,"type":15},{"name":13,"slug":14,"type":15},{"name":1200,"slug":1201,"type":15},"GitHub","github","2026-07-27T06:06:12.278222",{"slug":1204,"name":1204,"fn":1205,"description":1206,"org":1207,"tags":1208,"stars":1177,"repoUrl":1178,"updatedAt":1216},"mcore-create-issue","investigate CI failures and create issues","Investigate a failing GitHub Actions run or job and create a GitHub issue for the failure.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1209,1212,1213],{"name":1210,"slug":1211,"type":15},"Debugging","debugging",{"name":1200,"slug":1201,"type":15},{"name":1214,"slug":1215,"type":15},"Triage","triage","2026-07-14T05:25:57.442089",{"slug":1218,"name":1218,"fn":1219,"description":1220,"org":1221,"tags":1222,"stars":1177,"repoUrl":1178,"updatedAt":1229},"mcore-linting-and-formatting","lint and format Megatron-LM code","Linting and formatting for Megatron-LM. Covers running autoformat.sh, tools (ruff, black, isort, pylint, mypy), and code style rules.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1223,1226],{"name":1224,"slug":1225,"type":15},"Best Practices","best-practices",{"name":1227,"slug":1228,"type":15},"Code Analysis","code-analysis","2026-07-14T05:25:56.18433",{"slug":1231,"name":1231,"fn":1232,"description":1233,"org":1234,"tags":1235,"stars":1177,"repoUrl":1178,"updatedAt":1243},"mcore-migrate-gpt-to-hybrid","migrate Megatron-LM models to HybridModel","Migration guide for moving Megatron Core GPTModel checkpoints, model providers, training commands, and layer mappings to HybridModel.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1236,1239,1242],{"name":1237,"slug":1238,"type":15},"Machine Learning","machine-learning",{"name":1240,"slug":1241,"type":15},"Migration","migration",{"name":9,"slug":8,"type":15},"2026-07-17T06:07:11.777011",{"slug":1245,"name":1245,"fn":1246,"description":1247,"org":1248,"tags":1249,"stars":1177,"repoUrl":1178,"updatedAt":1256},"mcore-onboard-gb200-1node-tests","onboard functional tests for GB200","Onboard 1-node GitHub MR functional tests for GB200 from existing mr-scoped 2-node tests.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1250,1253],{"name":1251,"slug":1252,"type":15},"QA","qa",{"name":1254,"slug":1255,"type":15},"Testing","testing","2026-07-14T05:25:53.673039",{"slug":1258,"name":1258,"fn":1259,"description":1260,"org":1261,"tags":1262,"stars":1177,"repoUrl":1178,"updatedAt":1267},"mcore-run-on-slurm","launch distributed training jobs on SLURM","How to launch distributed Megatron-LM training jobs on a SLURM cluster. Covers a minimal sbatch skeleton, environment-variable setup for torch.distributed.run, CUDA_DEVICE_MAX_CONNECTIONS rules across hardware and parallelism modes, container conventions, monitoring, and per-rank failure diagnosis.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1263,1264],{"name":13,"slug":14,"type":15},{"name":1265,"slug":1266,"type":15},"Infrastructure","infrastructure","2026-07-14T05:25:49.362534",{"slug":1269,"name":1269,"fn":1270,"description":1271,"org":1272,"tags":1273,"stars":1177,"repoUrl":1178,"updatedAt":1281},"mcore-split-pr","split pull requests to reduce review load","Split a PR into multiple PRs to reduce the number of required CODEOWNERS reviewer groups.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1274,1277,1278],{"name":1275,"slug":1276,"type":15},"Code Review","code-review",{"name":1200,"slug":1201,"type":15},{"name":1279,"slug":1280,"type":15},"Pull Requests","pull-requests","2026-07-14T05:26:01.226578",{"slug":1283,"name":1283,"fn":1284,"description":1285,"org":1286,"tags":1287,"stars":1177,"repoUrl":1178,"updatedAt":1290},"mcore-testing","run and manage Megatron-LM tests","Test system for Megatron-LM. Covers test layout, recipe YAML structure, adding and running unit and functional tests, golden values, marker filters, and CI parity.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1288,1289],{"name":1251,"slug":1252,"type":15},{"name":1254,"slug":1255,"type":15},"2026-07-14T05:25:54.928983",{"slug":1292,"name":1292,"fn":1293,"description":1294,"org":1295,"tags":1296,"stars":1177,"repoUrl":1178,"updatedAt":1301},"nightly-sync","manage nightly main-to-dev sync workflows","Domain knowledge for the nightly main-to-dev sync workflow. Covers merge strategy, CI architecture, failure investigation, and known issues.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1297,1300],{"name":1298,"slug":1299,"type":15},"Automation","automation",{"name":1187,"slug":1188,"type":15},"2026-07-30T05:29:03.275638",496,{"items":1304,"total":1399},[1305,1322,1332,1346,1356,1371,1385],{"slug":1306,"name":1306,"fn":1307,"description":1308,"org":1309,"tags":1310,"stars":20,"repoUrl":21,"updatedAt":1321},"accelerated-computing-cudf","accelerate data processing with cuDF","Official NVIDIA-authored guidance for NVIDIA cuDF GPU DataFrames, pandas acceleration, dask-cuDF, ETL, joins, groupby, CSV\u002FParquet I\u002FO, nullable semantics, and multi-GPU DataFrame workloads.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1311,1314,1317,1318],{"name":1312,"slug":1313,"type":15},"Data Analysis","data-analysis",{"name":1315,"slug":1316,"type":15},"Data Engineering","data-engineering",{"name":9,"slug":8,"type":15},{"name":1319,"slug":1320,"type":15},"Performance","performance","2026-07-14T05:28:43.176466",{"slug":1323,"name":1323,"fn":1324,"description":1325,"org":1326,"tags":1327,"stars":20,"repoUrl":21,"updatedAt":1331},"aiq-deploy","deploy and manage NVIDIA AI-Q infrastructure","Use when asked to install, deploy, run, validate, troubleshoot, or stop NVIDIA AI-Q Blueprint infrastructure.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1328,1329,1330],{"name":13,"slug":14,"type":15},{"name":1265,"slug":1266,"type":15},{"name":9,"slug":8,"type":15},"2026-07-14T05:29:06.667109",{"slug":1333,"name":1333,"fn":1334,"description":1335,"org":1336,"tags":1337,"stars":20,"repoUrl":21,"updatedAt":1345},"aiq-research","conduct deep research with AI-Q","Use when asked to run deep research or AI-Q research through a reachable NVIDIA AI-Q Blueprint backend.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1338,1341,1342],{"name":1339,"slug":1340,"type":15},"Agents","agents",{"name":9,"slug":8,"type":15},{"name":1343,"slug":1344,"type":15},"Research","research","2026-07-14T05:28:06.816956",{"slug":1347,"name":1347,"fn":1348,"description":1349,"org":1350,"tags":1351,"stars":20,"repoUrl":21,"updatedAt":1355},"amc-run-sample-calibration","run AMC sample dataset calibration","Run end-to-end calibration on the shipped sample dataset (sdg_08_2_sample_data_010926.zip) against a running AMC microservice. Use when user says 'test sample dataset', 'run sample calibration', 'verify AMC install', or 'launch and test'.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1352,1353,1354],{"name":1312,"slug":1313,"type":15},{"name":9,"slug":8,"type":15},{"name":1254,"slug":1255,"type":15},"2026-07-17T05:29:03.913266",{"slug":1357,"name":1357,"fn":1358,"description":1359,"org":1360,"tags":1361,"stars":20,"repoUrl":21,"updatedAt":1370},"amc-run-video-calibration","calibrate video datasets with AutoMagicCalib","Calibrate a new dataset from pre-recorded video files via the AutoMagicCalib REST API. Use when user has local MP4s and says 'calibrate my videos', 'run AMC on these videos', or similar. For RTSP\u002Flive streams, use amc-run-rtsp-calibration instead.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1362,1363,1366,1367],{"name":1298,"slug":1299,"type":15},{"name":1364,"slug":1365,"type":15},"Imaging","imaging",{"name":9,"slug":8,"type":15},{"name":1368,"slug":1369,"type":15},"Video","video","2026-07-17T05:28:53.905004",{"slug":1372,"name":1372,"fn":1373,"description":1374,"org":1375,"tags":1376,"stars":20,"repoUrl":21,"updatedAt":1384},"amc-setup-calibration-stack","deploy AutoMagicCalib microservice with Docker","Launch AutoMagicCalib microservice and web UI from NGC release images via Docker Compose. Use when user says 'deploy auto calibration', 'launch auto calibration', 'launch AMC', 'start MS+UI', or 'set up auto-magic-calib'. Requires NGC API key.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1377,1378,1380,1381],{"name":13,"slug":14,"type":15},{"name":1379,"slug":235,"type":15},"Docker",{"name":9,"slug":8,"type":15},{"name":1382,"slug":1383,"type":15},"Operations","operations","2026-07-17T05:28:56.913999",{"slug":1386,"name":1386,"fn":1387,"description":1388,"org":1389,"tags":1390,"stars":20,"repoUrl":21,"updatedAt":1398},"cudaq-guide","develop quantum applications with CUDA-Q","CUDA-Q onboarding guide for installation, test programs, GPU simulation, QPU hardware, and quantum applications.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1391,1392,1395],{"name":9,"slug":8,"type":15},{"name":1393,"slug":1394,"type":15},"Quantum Computing","quantum-computing",{"name":1396,"slug":1397,"type":15},"Simulation","simulation","2026-07-14T05:26:58.898253",305]