[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-nvidia-sop-build":3,"mdc-oh0wo4-key":37,"related-repo-nvidia-sop-build":1993,"related-org-nvidia-sop-build":2086},{"slug":4,"name":4,"fn":5,"description":6,"org":7,"tags":11,"stars":26,"repoUrl":27,"updatedAt":28,"license":29,"forks":30,"topics":31,"repo":32,"sourceUrl":35,"mdContent":36},"sop-build","orchestrate end-to-end SOP monitoring pipelines","Orchestrate the end-to-end SOP pipeline, including preflight prerequisite checks, verifying models and downloading assets, generating the DeepStream SOP microservice with RTSP output, evaluating the microservice, and building, deploying, and testing the VSS SOP blueprint. Use when asked to run the full SOP pipeline, set up the SOP pipeline from scratch, execute preflight checks, verify models, download assets, generate the SOP microservice, evaluate the microservice, build the VSS blueprint, deploy the VSS blueprint, test the VSS blueprint, or manage the complete build-evaluate-deploy-test cycle.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},"nvidia","NVIDIA","https:\u002F\u002Fpexgzepcugksgbtrxkhf.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Forg-logos\u002Fnvidia.png",[12,16,19,20,23],{"name":13,"slug":14,"type":15},"Monitoring","monitoring","tag",{"name":17,"slug":18,"type":15},"Automation","automation",{"name":9,"slug":8,"type":15},{"name":21,"slug":22,"type":15},"Engineering","engineering",{"name":24,"slug":25,"type":15},"Video","video",39,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fsop-monitoring-blueprints","2026-07-14T05:32:48.503678","CC-BY-4.0 AND Apache-2.0",12,[],{"repoUrl":27,"stars":26,"forks":30,"topics":33,"description":34},[],"Industrial SOP Monitoring Blueprints for Training & Inference","https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fsop-monitoring-blueprints\u002Ftree\u002FHEAD\u002Fagentic\u002Fvss-sop-skills\u002Fsop-build","---\nname: sop-build\ndescription: >-\n  Orchestrate the end-to-end SOP pipeline, including preflight prerequisite checks, verifying models and downloading assets,\n  generating the DeepStream SOP microservice with RTSP output, evaluating the microservice, and building,\n  deploying, and testing the VSS SOP blueprint. Use when asked to run the full SOP pipeline, set up the SOP\n  pipeline from scratch, execute preflight checks, verify models, download assets, generate the SOP microservice,\n  evaluate the microservice, build the VSS blueprint, deploy the VSS blueprint, test the VSS blueprint, or\n  manage the complete build-evaluate-deploy-test cycle.\nowner: NVIDIA\nservice: vss-sop\nversion: 1.0.0\nlicense: CC-BY-4.0 AND Apache-2.0\nreviewed: 2026-06-23\nmetadata:\n  openclaw: { \"emoji\": \"🚀\", \"os\": [\"linux\"] }\n  author: \"Quan Vu \u003Cqvu@nvidia.com>\"\n  tags: [\"sop\", \"build\", \"pipeline\"]\n---\n\n# Full SOP Pipeline — Build, Evaluate, Deploy, Test\n\nOrchestrates the complete SOP pipeline from asset download through deployment validation. Each phase must succeed before proceeding to the next.\n\n## Overview\n\n- Set up the entire SOP pipeline from scratch on a new machine\n- Re-run the full pipeline after upstream changes\n- Rebuild and redeploy after ds-sop-skills or vss-sop-build changes\n\n## Prerequisites\n\n| Requirement | Detail |\n|---|---|\n| GPU | NVIDIA H100 \u002F H200 \u002F A100 with >= 80 GB VRAM |\n| Driver | NVIDIA 580+ with CUDA 13 |\n| Software | Docker with BuildKit, NVIDIA Container Toolkit 1.18.1, `ngc` CLI |\n| Secret key | `\u003Cbp-repo>\u002F.secret\u002Fngc_api_key.txt` must contain a valid NGC API key |\n| ds-sop-skills | `..\u002Fds-sop-skills\u002F` directory with `deepstream-sop\u002FSKILL.md` and prompt files |\n\n## Usage\n\n| Phase | What | Key Script \u002F Skill |\n|---|---|---|\n| Pre | Prerequisites Check & Auto-Install (run first) | `vss-sop-deploy\u002Fscripts\u002Fpreflight_check.sh --fix` |\n| 0 | Verify Models & Download Assets | `vss-sop-deploy\u002Fscripts\u002Fdownload_assets.sh` |\n| 1 | Generate SOP Microservice | `..\u002Fds-sop-skills\u002Fexample_sop_prompt.md` |\n| 2 | Evaluate SOP Microservice | `..\u002Fds-sop-skills\u002Feval_sop_prompt.md` |\n| 3 | Build VSS-SOP Blueprint | **vss-sop-build** skill |\n| 4 | Deploy VSS-SOP Blueprint | **vss-sop-deploy** skill |\n| 5 | Test VSS-SOP Blueprint | **vss-sop-test** skill |\n\n---\n\n### Phase Pre — Prerequisites Check & Auto-Install\n\nRun this **first**, before any other phase. It checks secret key files, GPU drivers, Docker, NVIDIA Container Toolkit, and NGC configuration, and can automatically install or configure missing prerequisites (NVIDIA Driver 580, CUDA Toolkit 13, Docker, Docker Compose, NVIDIA Container Toolkit, NGC CLI, and NGC CLI configuration).\n\n```bash\ncd \u003Cbp-repo>\n# Run checks and auto-fix\u002Finstall any missing components\n.\u002Fagentic\u002Fvss-sop-skills\u002Fvss-sop-deploy\u002Fscripts\u002Fpreflight_check.sh --bp-repo . --fix\n```\n\nTo run the checks only, without modifying your system:\n\n```bash\ncd \u003Cbp-repo>\n.\u002Fagentic\u002Fvss-sop-skills\u002Fvss-sop-deploy\u002Fscripts\u002Fpreflight_check.sh --bp-repo .\n```\n\n**Verify:** the script reports all prerequisites satisfied (secret keys, NVIDIA driver, Docker, NVIDIA Container Toolkit, NGC CLI & config). If any manual check fails, resolve it per the printed advice before proceeding. See **vss-sop-deploy** Phase 0 for reference docs.\n\n---\n\n### Phase 0 — Verify Models & Download Assets\n\nFor optimal accuracy, you must retrain\u002Ffine-tune the models, which can be done using the SOP Training Blueprint. After training, move model and config to `\u002Fopt\u002Fmodels\u002F...` and `\u002Fopt\u002Fsop\u002F...` directories.\n\n```bash\ncd \u003Cbp-repo>\n.\u002Fagentic\u002Fvss-sop-skills\u002Fvss-sop-deploy\u002Fscripts\u002Fdownload_assets.sh --bp-repo .\n```\n\nThis performs:\n1. Checks that the required trained model and config files exist under `\u002Fopt\u002Fmodels` and `\u002Fopt\u002Fsop`. If not, prompts the user to retrain\u002Ffine-tune the models using the SOP Training Blueprint.\n2. Downloads and re-encodes the sample RTSP video to H.264 at 30 FPS.\n3. Sets up the `ds-sop` cache directory and data log directories.\n\n**Verify:** `\u002Fopt\u002Fmodels\u002Fvlm\u002Fcheckpoint\u002F` contains VLM weights, `\u002Fopt\u002Fmodels\u002Fgbed_models\u002Fddm\u002Fcheckpoint.pth.tar` exists, `\u002Fopt\u002Fsop\u002Fconfigs\u002Factions.json` and `\u002Fopt\u002Fsop\u002Fconfigs\u002Fvlm_prompts.txt` exist.\n\n---\n\n### Phase 1 — Generate SOP Microservice\n\nGenerate the DeepStream SOP microservice with RTSP streaming output into `..\u002Fds_sop_microservice`. Skip if `..\u002Fds_sop_microservice\u002F` already exists with the expected layout.\n\nThe generation prompt **must** include the RTSP streaming output feature request (..\u002Fds-sop-skills § 18), which the VSS-SOP blueprint requires:\n\n```\nPlease follow instructions in ..\u002Fds-sop-skills\u002Fexample_sop_prompt.md to generate a SOP microservice with rtsp streaming output feature ($18) in folder ..\u002Fds_sop_microservice\n```\n\nThis follows `..\u002Fds-sop-skills\u002Fdeepstream-sop\u002FSKILL.md` to generate:\n- FastAPI microservice with `\u002Fv1\u002Fchat\u002Fcompletions` SSE endpoint\n- DeepStream GEBD pipeline with DDM-Net via Triton CAPI\n- VLM inference (Cosmos Reason) for action classification\n- SOP sequence checker with compliance monitoring\n- Kafka messaging (NvProto + JSON schema)\n- Docker build file and deploy compose\n- **RTSP streaming output** (`RTSPStreamingServer`, `tee1`-tapped encode branch, `SW_ENCODER`\u002F`ENABLE_RTSP_OUTPUT`\u002F`RTSP_PORT`)\n\n**Verify:** `..\u002Fds_sop_microservice\u002F` contains `docker\u002FDocker.build`, `deploy\u002Fcompose.yaml`, `nvds_action_detector\u002F{api_server,ds_3d_action_pipeline,ds_sop_process,vllm_inference}.py`.\n\n---\n\n### Phase 2 — Evaluate SOP Microservice\n\nEvaluate the generated microservice to catch issues before the full blueprint build.\n\n```\nPlease follow ..\u002Fds-sop-skills\u002Feval_sop_prompt.md to evaluate ..\u002Fds_sop_microservice with rtsp streaming output feature ($18)\n```\n\nUse the following environment settings:\n\n```\nVLLM_MODEL_PATH=\"\u002Fopt\u002Fmodels\u002Fvlm\u002Fcheckpoint\u002F\"\nDDM_MODEL_PATH=\"\u002Fopt\u002Fmodels\u002Fgbed_models\u002Fddm\u002Fcheckpoint.pth.tar\"\nMODEL_ROOT_DIR=\u002Fopt\u002Fmodels\nACTION_CONFIG_PATH=\u002Fopt\u002Fsop\u002Fconfigs\u002Factions.json\nVLM_PROMPT_PATH=\u002Fopt\u002Fsop\u002Fconfigs\u002Fvlm_prompts.txt\nHOST_CACHE=$HOME\u002F.cache\u002Fds-sop\nTEST_VIDEO_PATH=.\u002Fsop-resources\u002Fsop-server-fan-installation-data_v1.0-260213\u002Fserver_fan\u002Fraw\u002FInstall_1_h264_30fps.mp4\nUSER_ID=0\nGROUP_ID=0\n```\n\nThe Pylon SDK binary is at `..\u002Fds_sop_microservice\u002Fbinaries\u002Fpylon-25.10.2_linux-x86_64_setup.tar.gz`.\n\nThe evaluation follows `..\u002Fds-sop-skills\u002Fdeepstream-sop\u002Freferences\u002Fskill_12_evaluation_workflow.md` and performs:\n1. Static validation of generated source.\n2. Docker image build.\n3. Service launch with `deploy\u002F.env`.\n4. API unit tests.\n5. File-video API evaluation.\n6. Latency, camera, or Kafka evaluation as applicable.\n\n**Fix any issues found during evaluation before proceeding.** Record commands, results, timings, and fixes in the evaluation report.\n\n---\n\n### Phase 3 — Build VSS-SOP Blueprint (vss-sop-build)\n\nRun the **vss-sop-build** skill from the blueprint repo root. This scaffolds the SOP blueprint layer on top of VSS 3.1:\n\n```bash\ncd \u003Cbp-repo>\n# Follow vss-sop-build SKILL.md Quick Start steps 1–10\n```\n\nKey stages:\n1. Clone upstream VSS 3.1.0 repository.\n2. Copy all reference files verbatim.\n3. Copy and modify foundational services for SOP.\n4. Copy and modify VIOS for SOP.\n5. Copy agents from upstream and apply SOP profile.\n6. Copy NIM from upstream with SOP conventions.\n7. Build `ds-sop:1.0.0` image from the generated `..\u002Fds_sop_microservice`.\n8. Verify the build with `verify_build.sh`.\n\nSee [vss-sop-build SKILL.md](..\u002Fvss-sop-build\u002FSKILL.md) for full details.\n\n---\n\n### Phase 4 — Deploy VSS-SOP Blueprint (vss-sop-deploy)\n\nRun the **vss-sop-deploy** skill from the blueprint repo root:\n\n```bash\ncd \u003Cbp-repo>\n# Follow vss-sop-deploy SKILL.md Phases 0–5\n```\n\nDeployment phases:\n- **Phase 0:** Prerequisites check and auto-install (`preflight_check.sh --bp-repo . --fix`) — already run in **Phase Pre** above; re-runs as an idempotent safety check.\n- **Phase 1:** Download models & assets (already done in Phase 0 above — will skip if present).\n- **Phase 2:** Start RTSP server simulation (`start_rtsp_server.sh --bp-repo .`).\n- **Phase 3:** Build DS-SOP Docker image & verify components (already built in Phase 3 above — verify only).\n- **Phase 4:** Configure & deploy the blueprint (`configure_blueprint.sh`, `deploy.sh`).\n- **Phase 5:** Test with RTSP stream (`test_rtsp.sh --bp-repo .`).\n\nSee [vss-sop-deploy SKILL.md](..\u002Fvss-sop-deploy\u002FSKILL.md) for full details.\n\n---\n\n### Phase 5 — Test VSS-SOP Blueprint (vss-sop-test)\n\nRun the **vss-sop-test** skill from the blueprint repo root:\n\n```bash\ncd \u003Cbp-repo>\npython agentic\u002Fvss-sop-skills\u002Fvss-sop-test\u002Fscripts\u002Fvss_sop_test.py \\\n  --bp-repo . \\\n  --env-file deployments\u002Fsop\u002F.env\n```\n\nTest phases:\n- **Phase 1:** Service health — all required containers running.\n- **Phase 2:** ELK data pipeline — Elasticsearch, indices, VLM messages, Kibana fields.\n- **Phase 3:** VIOS recording & livestream — sensors, streams, recording, livestream.\n- **Phase 4:** VSS Agent end-to-end — MCP, LLM, VLM, snapshot, video, report generation.\n\nSee [vss-sop-test SKILL.md](..\u002Fvss-sop-test\u002FSKILL.md) for full details.\n\n---\n\n## Error Handling\n\n### Pass Criteria and Verification\n\nEnsure all unit tests, build checks, deploy checks, and test phases pass with zero failures.\n- Do not classify any failure as environmental, intermittent, infrastructure, or acceptable. Treat every failure as a bug to be fixed.\n- Do not proceed to the next phase if the current phase has any failing test. Fix all failures first.\n- Do not declare completion if any test is failing. A result like \"36\u002F38 checks pass\" is a failure, not a success.\n- If a test times out, fix the timeout or the underlying latency rather than dismissing it.\n- If stuck on the same failure after three fix attempts, stop and ask the user for help. Do not skip the failure.\n\n### Auto-Debug and Fix Loop\n\nWhen any phase fails:\n1. Read the failure output and auto-debug hints.\n2. Collect container logs (`docker logs \u003Ccontainer> --tail 50`).\n3. Apply the fix (env, compose, patches, image rebuild, source regeneration).\n4. Re-run the failing phase to confirm the fix.\n5. Repeat until all checks in that phase pass.\n\nIf the root cause is in the generated DS-SOP source code, update `ds-sop-skills` references and regenerate. If the root cause is in the build\u002Fdeploy config, update `vss-sop-build` references.\n\n---\n\n## Examples\n\nAfter all phases complete with 100% pass rate, present a summary table:\n\n| Category | Status | Details |\n|---|---|---|\n| Phase Pre: Prerequisites | Pass\u002FFail | Driver, Docker, Toolkit, NGC config |\n| Phase 0: Assets | Pass\u002FFail | Models and configs at \u002Fopt |\n| Phase 1: Generate | Pass\u002FFail | ..\u002Fds_sop_microservice with RTSP |\n| Phase 2: Evaluate | Pass\u002FFail | Unit tests + API evaluation |\n| Phase 3: Build | Pass\u002FFail | Per-stage build status |\n| Phase 4: Deploy | Pass\u002FFail | Per-phase deploy status |\n| Phase 5: Test | Pass\u002FFail | Per-phase test status (must be 100%) |\n\nInclude:\n- Any `vss-sop-build` or `ds-sop-skills` files updated during debugging.\n- Service endpoints: VSS-UI, Kibana, VIOS-UI, Grafana.\n\nIf any test is still failing, the report must clearly state **\"Pipeline Incomplete — X failures remain\"** and list them. Do not present a partial success as completion.\n\n### Service Endpoints (after successful deploy)\n\n| Service | Endpoint |\n|---|---|\n| VSS-UI | `http:\u002F\u002F\u003CEXTERNAL_IP>:3000` |\n| Kibana | `http:\u002F\u002F\u003CEXTERNAL_IP>:5601\u002Fapp\u002Fhome#\u002F` |\n| VIOS-UI | `http:\u002F\u002F\u003CEXTERNAL_IP>:30888\u002Fvst\u002F#\u002Fdashboard` |\n| Grafana | `http:\u002F\u002F\u003CEXTERNAL_IP>:35000\u002F` |\n| Phoenix (Telemetry) | `http:\u002F\u002F\u003CEXTERNAL_IP>:6006\u002Fprojects` |\n\n---\n\n## License\n\nUse of this skill is governed by the [Creative Commons Attribution 4.0 International License (CC BY 4.0)](https:\u002F\u002Fcreativecommons.org\u002Flicenses\u002Fby\u002F4.0\u002Flegalcode.en) and the [Apache License, Version 2.0](https:\u002F\u002Fwww.apache.org\u002Flicenses\u002FLICENSE-2.0).\n",{"data":38,"body":52},{"name":4,"description":6,"owner":9,"service":39,"version":40,"license":29,"reviewed":41,"metadata":42},"vss-sop","1.0.0","2026-06-23",{"openclaw":43,"author":47,"tags":48},{"emoji":44,"os":45},"🚀",[46],"linux","Quan Vu \u003Cqvu@nvidia.com>",[49,50,51],"sop","build","pipeline",{"type":53,"children":54},"root",[55,64,70,77,97,103,225,231,414,418,425,437,518,523,569,585,588,594,615,661,666,708,748,751,757,778,790,800,813,903,940,943,949,954,963,968,977,989,1002,1042,1052,1055,1061,1072,1110,1115,1179,1193,1196,1202,1212,1250,1255,1362,1373,1376,1382,1391,1469,1474,1513,1524,1527,1533,1539,1544,1572,1578,1583,1618,1637,1640,1646,1651,1798,1803,1830,1842,1848,1955,1958,1964,1987],{"type":56,"tag":57,"props":58,"children":60},"element","h1",{"id":59},"full-sop-pipeline-build-evaluate-deploy-test",[61],{"type":62,"value":63},"text","Full SOP Pipeline — Build, Evaluate, Deploy, Test",{"type":56,"tag":65,"props":66,"children":67},"p",{},[68],{"type":62,"value":69},"Orchestrates the complete SOP pipeline from asset download through deployment validation. Each phase must succeed before proceeding to the next.",{"type":56,"tag":71,"props":72,"children":74},"h2",{"id":73},"overview",[75],{"type":62,"value":76},"Overview",{"type":56,"tag":78,"props":79,"children":80},"ul",{},[81,87,92],{"type":56,"tag":82,"props":83,"children":84},"li",{},[85],{"type":62,"value":86},"Set up the entire SOP pipeline from scratch on a new machine",{"type":56,"tag":82,"props":88,"children":89},{},[90],{"type":62,"value":91},"Re-run the full pipeline after upstream changes",{"type":56,"tag":82,"props":93,"children":94},{},[95],{"type":62,"value":96},"Rebuild and redeploy after ds-sop-skills or vss-sop-build changes",{"type":56,"tag":71,"props":98,"children":100},{"id":99},"prerequisites",[101],{"type":62,"value":102},"Prerequisites",{"type":56,"tag":104,"props":105,"children":106},"table",{},[107,126],{"type":56,"tag":108,"props":109,"children":110},"thead",{},[111],{"type":56,"tag":112,"props":113,"children":114},"tr",{},[115,121],{"type":56,"tag":116,"props":117,"children":118},"th",{},[119],{"type":62,"value":120},"Requirement",{"type":56,"tag":116,"props":122,"children":123},{},[124],{"type":62,"value":125},"Detail",{"type":56,"tag":127,"props":128,"children":129},"tbody",{},[130,144,157,179,198],{"type":56,"tag":112,"props":131,"children":132},{},[133,139],{"type":56,"tag":134,"props":135,"children":136},"td",{},[137],{"type":62,"value":138},"GPU",{"type":56,"tag":134,"props":140,"children":141},{},[142],{"type":62,"value":143},"NVIDIA H100 \u002F H200 \u002F A100 with >= 80 GB VRAM",{"type":56,"tag":112,"props":145,"children":146},{},[147,152],{"type":56,"tag":134,"props":148,"children":149},{},[150],{"type":62,"value":151},"Driver",{"type":56,"tag":134,"props":153,"children":154},{},[155],{"type":62,"value":156},"NVIDIA 580+ with CUDA 13",{"type":56,"tag":112,"props":158,"children":159},{},[160,165],{"type":56,"tag":134,"props":161,"children":162},{},[163],{"type":62,"value":164},"Software",{"type":56,"tag":134,"props":166,"children":167},{},[168,170,177],{"type":62,"value":169},"Docker with BuildKit, NVIDIA Container Toolkit 1.18.1, ",{"type":56,"tag":171,"props":172,"children":174},"code",{"className":173},[],[175],{"type":62,"value":176},"ngc",{"type":62,"value":178}," CLI",{"type":56,"tag":112,"props":180,"children":181},{},[182,187],{"type":56,"tag":134,"props":183,"children":184},{},[185],{"type":62,"value":186},"Secret key",{"type":56,"tag":134,"props":188,"children":189},{},[190,196],{"type":56,"tag":171,"props":191,"children":193},{"className":192},[],[194],{"type":62,"value":195},"\u003Cbp-repo>\u002F.secret\u002Fngc_api_key.txt",{"type":62,"value":197}," must contain a valid NGC API key",{"type":56,"tag":112,"props":199,"children":200},{},[201,206],{"type":56,"tag":134,"props":202,"children":203},{},[204],{"type":62,"value":205},"ds-sop-skills",{"type":56,"tag":134,"props":207,"children":208},{},[209,215,217,223],{"type":56,"tag":171,"props":210,"children":212},{"className":211},[],[213],{"type":62,"value":214},"..\u002Fds-sop-skills\u002F",{"type":62,"value":216}," directory with ",{"type":56,"tag":171,"props":218,"children":220},{"className":219},[],[221],{"type":62,"value":222},"deepstream-sop\u002FSKILL.md",{"type":62,"value":224}," and prompt files",{"type":56,"tag":71,"props":226,"children":228},{"id":227},"usage",[229],{"type":62,"value":230},"Usage",{"type":56,"tag":104,"props":232,"children":233},{},[234,255],{"type":56,"tag":108,"props":235,"children":236},{},[237],{"type":56,"tag":112,"props":238,"children":239},{},[240,245,250],{"type":56,"tag":116,"props":241,"children":242},{},[243],{"type":62,"value":244},"Phase",{"type":56,"tag":116,"props":246,"children":247},{},[248],{"type":62,"value":249},"What",{"type":56,"tag":116,"props":251,"children":252},{},[253],{"type":62,"value":254},"Key Script \u002F Skill",{"type":56,"tag":127,"props":256,"children":257},{},[258,280,302,324,346,370,392],{"type":56,"tag":112,"props":259,"children":260},{},[261,266,271],{"type":56,"tag":134,"props":262,"children":263},{},[264],{"type":62,"value":265},"Pre",{"type":56,"tag":134,"props":267,"children":268},{},[269],{"type":62,"value":270},"Prerequisites Check & Auto-Install (run first)",{"type":56,"tag":134,"props":272,"children":273},{},[274],{"type":56,"tag":171,"props":275,"children":277},{"className":276},[],[278],{"type":62,"value":279},"vss-sop-deploy\u002Fscripts\u002Fpreflight_check.sh --fix",{"type":56,"tag":112,"props":281,"children":282},{},[283,288,293],{"type":56,"tag":134,"props":284,"children":285},{},[286],{"type":62,"value":287},"0",{"type":56,"tag":134,"props":289,"children":290},{},[291],{"type":62,"value":292},"Verify Models & Download Assets",{"type":56,"tag":134,"props":294,"children":295},{},[296],{"type":56,"tag":171,"props":297,"children":299},{"className":298},[],[300],{"type":62,"value":301},"vss-sop-deploy\u002Fscripts\u002Fdownload_assets.sh",{"type":56,"tag":112,"props":303,"children":304},{},[305,310,315],{"type":56,"tag":134,"props":306,"children":307},{},[308],{"type":62,"value":309},"1",{"type":56,"tag":134,"props":311,"children":312},{},[313],{"type":62,"value":314},"Generate SOP Microservice",{"type":56,"tag":134,"props":316,"children":317},{},[318],{"type":56,"tag":171,"props":319,"children":321},{"className":320},[],[322],{"type":62,"value":323},"..\u002Fds-sop-skills\u002Fexample_sop_prompt.md",{"type":56,"tag":112,"props":325,"children":326},{},[327,332,337],{"type":56,"tag":134,"props":328,"children":329},{},[330],{"type":62,"value":331},"2",{"type":56,"tag":134,"props":333,"children":334},{},[335],{"type":62,"value":336},"Evaluate SOP Microservice",{"type":56,"tag":134,"props":338,"children":339},{},[340],{"type":56,"tag":171,"props":341,"children":343},{"className":342},[],[344],{"type":62,"value":345},"..\u002Fds-sop-skills\u002Feval_sop_prompt.md",{"type":56,"tag":112,"props":347,"children":348},{},[349,354,359],{"type":56,"tag":134,"props":350,"children":351},{},[352],{"type":62,"value":353},"3",{"type":56,"tag":134,"props":355,"children":356},{},[357],{"type":62,"value":358},"Build VSS-SOP Blueprint",{"type":56,"tag":134,"props":360,"children":361},{},[362,368],{"type":56,"tag":363,"props":364,"children":365},"strong",{},[366],{"type":62,"value":367},"vss-sop-build",{"type":62,"value":369}," skill",{"type":56,"tag":112,"props":371,"children":372},{},[373,378,383],{"type":56,"tag":134,"props":374,"children":375},{},[376],{"type":62,"value":377},"4",{"type":56,"tag":134,"props":379,"children":380},{},[381],{"type":62,"value":382},"Deploy VSS-SOP Blueprint",{"type":56,"tag":134,"props":384,"children":385},{},[386,391],{"type":56,"tag":363,"props":387,"children":388},{},[389],{"type":62,"value":390},"vss-sop-deploy",{"type":62,"value":369},{"type":56,"tag":112,"props":393,"children":394},{},[395,400,405],{"type":56,"tag":134,"props":396,"children":397},{},[398],{"type":62,"value":399},"5",{"type":56,"tag":134,"props":401,"children":402},{},[403],{"type":62,"value":404},"Test VSS-SOP Blueprint",{"type":56,"tag":134,"props":406,"children":407},{},[408,413],{"type":56,"tag":363,"props":409,"children":410},{},[411],{"type":62,"value":412},"vss-sop-test",{"type":62,"value":369},{"type":56,"tag":415,"props":416,"children":417},"hr",{},[],{"type":56,"tag":419,"props":420,"children":422},"h3",{"id":421},"phase-pre-prerequisites-check-auto-install",[423],{"type":62,"value":424},"Phase Pre — Prerequisites Check & Auto-Install",{"type":56,"tag":65,"props":426,"children":427},{},[428,430,435],{"type":62,"value":429},"Run this ",{"type":56,"tag":363,"props":431,"children":432},{},[433],{"type":62,"value":434},"first",{"type":62,"value":436},", before any other phase. It checks secret key files, GPU drivers, Docker, NVIDIA Container Toolkit, and NGC configuration, and can automatically install or configure missing prerequisites (NVIDIA Driver 580, CUDA Toolkit 13, Docker, Docker Compose, NVIDIA Container Toolkit, NGC CLI, and NGC CLI configuration).",{"type":56,"tag":438,"props":439,"children":444},"pre",{"className":440,"code":441,"language":442,"meta":443,"style":443},"language-bash shiki shiki-themes material-theme-lighter material-theme material-theme-palenight","cd \u003Cbp-repo>\n# Run checks and auto-fix\u002Finstall any missing components\n.\u002Fagentic\u002Fvss-sop-skills\u002Fvss-sop-deploy\u002Fscripts\u002Fpreflight_check.sh --bp-repo . --fix\n","bash","",[445],{"type":56,"tag":171,"props":446,"children":447},{"__ignoreMap":443},[448,483,493],{"type":56,"tag":449,"props":450,"children":453},"span",{"class":451,"line":452},"line",1,[454,460,466,472,478],{"type":56,"tag":449,"props":455,"children":457},{"style":456},"--shiki-light:#6182B8;--shiki-default:#82AAFF;--shiki-dark:#82AAFF",[458],{"type":62,"value":459},"cd",{"type":56,"tag":449,"props":461,"children":463},{"style":462},"--shiki-light:#39ADB5;--shiki-default:#89DDFF;--shiki-dark:#89DDFF",[464],{"type":62,"value":465}," \u003C",{"type":56,"tag":449,"props":467,"children":469},{"style":468},"--shiki-light:#91B859;--shiki-default:#C3E88D;--shiki-dark:#C3E88D",[470],{"type":62,"value":471},"bp-rep",{"type":56,"tag":449,"props":473,"children":475},{"style":474},"--shiki-light:#90A4AE;--shiki-default:#EEFFFF;--shiki-dark:#BABED8",[476],{"type":62,"value":477},"o",{"type":56,"tag":449,"props":479,"children":480},{"style":462},[481],{"type":62,"value":482},">\n",{"type":56,"tag":449,"props":484,"children":486},{"class":451,"line":485},2,[487],{"type":56,"tag":449,"props":488,"children":490},{"style":489},"--shiki-light:#90A4AE;--shiki-light-font-style:italic;--shiki-default:#546E7A;--shiki-default-font-style:italic;--shiki-dark:#676E95;--shiki-dark-font-style:italic",[491],{"type":62,"value":492},"# Run checks and auto-fix\u002Finstall any missing components\n",{"type":56,"tag":449,"props":494,"children":496},{"class":451,"line":495},3,[497,503,508,513],{"type":56,"tag":449,"props":498,"children":500},{"style":499},"--shiki-light:#E2931D;--shiki-default:#FFCB6B;--shiki-dark:#FFCB6B",[501],{"type":62,"value":502},".\u002Fagentic\u002Fvss-sop-skills\u002Fvss-sop-deploy\u002Fscripts\u002Fpreflight_check.sh",{"type":56,"tag":449,"props":504,"children":505},{"style":468},[506],{"type":62,"value":507}," --bp-repo",{"type":56,"tag":449,"props":509,"children":510},{"style":468},[511],{"type":62,"value":512}," .",{"type":56,"tag":449,"props":514,"children":515},{"style":468},[516],{"type":62,"value":517}," --fix\n",{"type":56,"tag":65,"props":519,"children":520},{},[521],{"type":62,"value":522},"To run the checks only, without modifying your system:",{"type":56,"tag":438,"props":524,"children":526},{"className":440,"code":525,"language":442,"meta":443,"style":443},"cd \u003Cbp-repo>\n.\u002Fagentic\u002Fvss-sop-skills\u002Fvss-sop-deploy\u002Fscripts\u002Fpreflight_check.sh --bp-repo .\n",[527],{"type":56,"tag":171,"props":528,"children":529},{"__ignoreMap":443},[530,553],{"type":56,"tag":449,"props":531,"children":532},{"class":451,"line":452},[533,537,541,545,549],{"type":56,"tag":449,"props":534,"children":535},{"style":456},[536],{"type":62,"value":459},{"type":56,"tag":449,"props":538,"children":539},{"style":462},[540],{"type":62,"value":465},{"type":56,"tag":449,"props":542,"children":543},{"style":468},[544],{"type":62,"value":471},{"type":56,"tag":449,"props":546,"children":547},{"style":474},[548],{"type":62,"value":477},{"type":56,"tag":449,"props":550,"children":551},{"style":462},[552],{"type":62,"value":482},{"type":56,"tag":449,"props":554,"children":555},{"class":451,"line":485},[556,560,564],{"type":56,"tag":449,"props":557,"children":558},{"style":499},[559],{"type":62,"value":502},{"type":56,"tag":449,"props":561,"children":562},{"style":468},[563],{"type":62,"value":507},{"type":56,"tag":449,"props":565,"children":566},{"style":468},[567],{"type":62,"value":568}," .\n",{"type":56,"tag":65,"props":570,"children":571},{},[572,577,579,583],{"type":56,"tag":363,"props":573,"children":574},{},[575],{"type":62,"value":576},"Verify:",{"type":62,"value":578}," the script reports all prerequisites satisfied (secret keys, NVIDIA driver, Docker, NVIDIA Container Toolkit, NGC CLI & config). If any manual check fails, resolve it per the printed advice before proceeding. See ",{"type":56,"tag":363,"props":580,"children":581},{},[582],{"type":62,"value":390},{"type":62,"value":584}," Phase 0 for reference docs.",{"type":56,"tag":415,"props":586,"children":587},{},[],{"type":56,"tag":419,"props":589,"children":591},{"id":590},"phase-0-verify-models-download-assets",[592],{"type":62,"value":593},"Phase 0 — Verify Models & Download Assets",{"type":56,"tag":65,"props":595,"children":596},{},[597,599,605,607,613],{"type":62,"value":598},"For optimal accuracy, you must retrain\u002Ffine-tune the models, which can be done using the SOP Training Blueprint. After training, move model and config to ",{"type":56,"tag":171,"props":600,"children":602},{"className":601},[],[603],{"type":62,"value":604},"\u002Fopt\u002Fmodels\u002F...",{"type":62,"value":606}," and ",{"type":56,"tag":171,"props":608,"children":610},{"className":609},[],[611],{"type":62,"value":612},"\u002Fopt\u002Fsop\u002F...",{"type":62,"value":614}," directories.",{"type":56,"tag":438,"props":616,"children":618},{"className":440,"code":617,"language":442,"meta":443,"style":443},"cd \u003Cbp-repo>\n.\u002Fagentic\u002Fvss-sop-skills\u002Fvss-sop-deploy\u002Fscripts\u002Fdownload_assets.sh --bp-repo .\n",[619],{"type":56,"tag":171,"props":620,"children":621},{"__ignoreMap":443},[622,645],{"type":56,"tag":449,"props":623,"children":624},{"class":451,"line":452},[625,629,633,637,641],{"type":56,"tag":449,"props":626,"children":627},{"style":456},[628],{"type":62,"value":459},{"type":56,"tag":449,"props":630,"children":631},{"style":462},[632],{"type":62,"value":465},{"type":56,"tag":449,"props":634,"children":635},{"style":468},[636],{"type":62,"value":471},{"type":56,"tag":449,"props":638,"children":639},{"style":474},[640],{"type":62,"value":477},{"type":56,"tag":449,"props":642,"children":643},{"style":462},[644],{"type":62,"value":482},{"type":56,"tag":449,"props":646,"children":647},{"class":451,"line":485},[648,653,657],{"type":56,"tag":449,"props":649,"children":650},{"style":499},[651],{"type":62,"value":652},".\u002Fagentic\u002Fvss-sop-skills\u002Fvss-sop-deploy\u002Fscripts\u002Fdownload_assets.sh",{"type":56,"tag":449,"props":654,"children":655},{"style":468},[656],{"type":62,"value":507},{"type":56,"tag":449,"props":658,"children":659},{"style":468},[660],{"type":62,"value":568},{"type":56,"tag":65,"props":662,"children":663},{},[664],{"type":62,"value":665},"This performs:",{"type":56,"tag":667,"props":668,"children":669},"ol",{},[670,690,695],{"type":56,"tag":82,"props":671,"children":672},{},[673,675,681,682,688],{"type":62,"value":674},"Checks that the required trained model and config files exist under ",{"type":56,"tag":171,"props":676,"children":678},{"className":677},[],[679],{"type":62,"value":680},"\u002Fopt\u002Fmodels",{"type":62,"value":606},{"type":56,"tag":171,"props":683,"children":685},{"className":684},[],[686],{"type":62,"value":687},"\u002Fopt\u002Fsop",{"type":62,"value":689},". If not, prompts the user to retrain\u002Ffine-tune the models using the SOP Training Blueprint.",{"type":56,"tag":82,"props":691,"children":692},{},[693],{"type":62,"value":694},"Downloads and re-encodes the sample RTSP video to H.264 at 30 FPS.",{"type":56,"tag":82,"props":696,"children":697},{},[698,700,706],{"type":62,"value":699},"Sets up the ",{"type":56,"tag":171,"props":701,"children":703},{"className":702},[],[704],{"type":62,"value":705},"ds-sop",{"type":62,"value":707}," cache directory and data log directories.",{"type":56,"tag":65,"props":709,"children":710},{},[711,715,717,723,725,731,733,739,740,746],{"type":56,"tag":363,"props":712,"children":713},{},[714],{"type":62,"value":576},{"type":62,"value":716}," ",{"type":56,"tag":171,"props":718,"children":720},{"className":719},[],[721],{"type":62,"value":722},"\u002Fopt\u002Fmodels\u002Fvlm\u002Fcheckpoint\u002F",{"type":62,"value":724}," contains VLM weights, ",{"type":56,"tag":171,"props":726,"children":728},{"className":727},[],[729],{"type":62,"value":730},"\u002Fopt\u002Fmodels\u002Fgbed_models\u002Fddm\u002Fcheckpoint.pth.tar",{"type":62,"value":732}," exists, ",{"type":56,"tag":171,"props":734,"children":736},{"className":735},[],[737],{"type":62,"value":738},"\u002Fopt\u002Fsop\u002Fconfigs\u002Factions.json",{"type":62,"value":606},{"type":56,"tag":171,"props":741,"children":743},{"className":742},[],[744],{"type":62,"value":745},"\u002Fopt\u002Fsop\u002Fconfigs\u002Fvlm_prompts.txt",{"type":62,"value":747}," exist.",{"type":56,"tag":415,"props":749,"children":750},{},[],{"type":56,"tag":419,"props":752,"children":754},{"id":753},"phase-1-generate-sop-microservice",[755],{"type":62,"value":756},"Phase 1 — Generate SOP Microservice",{"type":56,"tag":65,"props":758,"children":759},{},[760,762,768,770,776],{"type":62,"value":761},"Generate the DeepStream SOP microservice with RTSP streaming output into ",{"type":56,"tag":171,"props":763,"children":765},{"className":764},[],[766],{"type":62,"value":767},"..\u002Fds_sop_microservice",{"type":62,"value":769},". Skip if ",{"type":56,"tag":171,"props":771,"children":773},{"className":772},[],[774],{"type":62,"value":775},"..\u002Fds_sop_microservice\u002F",{"type":62,"value":777}," already exists with the expected layout.",{"type":56,"tag":65,"props":779,"children":780},{},[781,783,788],{"type":62,"value":782},"The generation prompt ",{"type":56,"tag":363,"props":784,"children":785},{},[786],{"type":62,"value":787},"must",{"type":62,"value":789}," include the RTSP streaming output feature request (..\u002Fds-sop-skills § 18), which the VSS-SOP blueprint requires:",{"type":56,"tag":438,"props":791,"children":795},{"className":792,"code":794,"language":62},[793],"language-text","Please follow instructions in ..\u002Fds-sop-skills\u002Fexample_sop_prompt.md to generate a SOP microservice with rtsp streaming output feature ($18) in folder ..\u002Fds_sop_microservice\n",[796],{"type":56,"tag":171,"props":797,"children":798},{"__ignoreMap":443},[799],{"type":62,"value":794},{"type":56,"tag":65,"props":801,"children":802},{},[803,805,811],{"type":62,"value":804},"This follows ",{"type":56,"tag":171,"props":806,"children":808},{"className":807},[],[809],{"type":62,"value":810},"..\u002Fds-sop-skills\u002Fdeepstream-sop\u002FSKILL.md",{"type":62,"value":812}," to generate:",{"type":56,"tag":78,"props":814,"children":815},{},[816,829,834,839,844,849,854],{"type":56,"tag":82,"props":817,"children":818},{},[819,821,827],{"type":62,"value":820},"FastAPI microservice with ",{"type":56,"tag":171,"props":822,"children":824},{"className":823},[],[825],{"type":62,"value":826},"\u002Fv1\u002Fchat\u002Fcompletions",{"type":62,"value":828}," SSE endpoint",{"type":56,"tag":82,"props":830,"children":831},{},[832],{"type":62,"value":833},"DeepStream GEBD pipeline with DDM-Net via Triton CAPI",{"type":56,"tag":82,"props":835,"children":836},{},[837],{"type":62,"value":838},"VLM inference (Cosmos Reason) for action classification",{"type":56,"tag":82,"props":840,"children":841},{},[842],{"type":62,"value":843},"SOP sequence checker with compliance monitoring",{"type":56,"tag":82,"props":845,"children":846},{},[847],{"type":62,"value":848},"Kafka messaging (NvProto + JSON schema)",{"type":56,"tag":82,"props":850,"children":851},{},[852],{"type":62,"value":853},"Docker build file and deploy compose",{"type":56,"tag":82,"props":855,"children":856},{},[857,862,864,870,872,878,880,886,888,894,895,901],{"type":56,"tag":363,"props":858,"children":859},{},[860],{"type":62,"value":861},"RTSP streaming output",{"type":62,"value":863}," (",{"type":56,"tag":171,"props":865,"children":867},{"className":866},[],[868],{"type":62,"value":869},"RTSPStreamingServer",{"type":62,"value":871},", ",{"type":56,"tag":171,"props":873,"children":875},{"className":874},[],[876],{"type":62,"value":877},"tee1",{"type":62,"value":879},"-tapped encode branch, ",{"type":56,"tag":171,"props":881,"children":883},{"className":882},[],[884],{"type":62,"value":885},"SW_ENCODER",{"type":62,"value":887},"\u002F",{"type":56,"tag":171,"props":889,"children":891},{"className":890},[],[892],{"type":62,"value":893},"ENABLE_RTSP_OUTPUT",{"type":62,"value":887},{"type":56,"tag":171,"props":896,"children":898},{"className":897},[],[899],{"type":62,"value":900},"RTSP_PORT",{"type":62,"value":902},")",{"type":56,"tag":65,"props":904,"children":905},{},[906,910,911,916,918,924,925,931,932,938],{"type":56,"tag":363,"props":907,"children":908},{},[909],{"type":62,"value":576},{"type":62,"value":716},{"type":56,"tag":171,"props":912,"children":914},{"className":913},[],[915],{"type":62,"value":775},{"type":62,"value":917}," contains ",{"type":56,"tag":171,"props":919,"children":921},{"className":920},[],[922],{"type":62,"value":923},"docker\u002FDocker.build",{"type":62,"value":871},{"type":56,"tag":171,"props":926,"children":928},{"className":927},[],[929],{"type":62,"value":930},"deploy\u002Fcompose.yaml",{"type":62,"value":871},{"type":56,"tag":171,"props":933,"children":935},{"className":934},[],[936],{"type":62,"value":937},"nvds_action_detector\u002F{api_server,ds_3d_action_pipeline,ds_sop_process,vllm_inference}.py",{"type":62,"value":939},".",{"type":56,"tag":415,"props":941,"children":942},{},[],{"type":56,"tag":419,"props":944,"children":946},{"id":945},"phase-2-evaluate-sop-microservice",[947],{"type":62,"value":948},"Phase 2 — Evaluate SOP Microservice",{"type":56,"tag":65,"props":950,"children":951},{},[952],{"type":62,"value":953},"Evaluate the generated microservice to catch issues before the full blueprint build.",{"type":56,"tag":438,"props":955,"children":958},{"className":956,"code":957,"language":62},[793],"Please follow ..\u002Fds-sop-skills\u002Feval_sop_prompt.md to evaluate ..\u002Fds_sop_microservice with rtsp streaming output feature ($18)\n",[959],{"type":56,"tag":171,"props":960,"children":961},{"__ignoreMap":443},[962],{"type":62,"value":957},{"type":56,"tag":65,"props":964,"children":965},{},[966],{"type":62,"value":967},"Use the following environment settings:",{"type":56,"tag":438,"props":969,"children":972},{"className":970,"code":971,"language":62},[793],"VLLM_MODEL_PATH=\"\u002Fopt\u002Fmodels\u002Fvlm\u002Fcheckpoint\u002F\"\nDDM_MODEL_PATH=\"\u002Fopt\u002Fmodels\u002Fgbed_models\u002Fddm\u002Fcheckpoint.pth.tar\"\nMODEL_ROOT_DIR=\u002Fopt\u002Fmodels\nACTION_CONFIG_PATH=\u002Fopt\u002Fsop\u002Fconfigs\u002Factions.json\nVLM_PROMPT_PATH=\u002Fopt\u002Fsop\u002Fconfigs\u002Fvlm_prompts.txt\nHOST_CACHE=$HOME\u002F.cache\u002Fds-sop\nTEST_VIDEO_PATH=.\u002Fsop-resources\u002Fsop-server-fan-installation-data_v1.0-260213\u002Fserver_fan\u002Fraw\u002FInstall_1_h264_30fps.mp4\nUSER_ID=0\nGROUP_ID=0\n",[973],{"type":56,"tag":171,"props":974,"children":975},{"__ignoreMap":443},[976],{"type":62,"value":971},{"type":56,"tag":65,"props":978,"children":979},{},[980,982,988],{"type":62,"value":981},"The Pylon SDK binary is at ",{"type":56,"tag":171,"props":983,"children":985},{"className":984},[],[986],{"type":62,"value":987},"..\u002Fds_sop_microservice\u002Fbinaries\u002Fpylon-25.10.2_linux-x86_64_setup.tar.gz",{"type":62,"value":939},{"type":56,"tag":65,"props":990,"children":991},{},[992,994,1000],{"type":62,"value":993},"The evaluation follows ",{"type":56,"tag":171,"props":995,"children":997},{"className":996},[],[998],{"type":62,"value":999},"..\u002Fds-sop-skills\u002Fdeepstream-sop\u002Freferences\u002Fskill_12_evaluation_workflow.md",{"type":62,"value":1001}," and performs:",{"type":56,"tag":667,"props":1003,"children":1004},{},[1005,1010,1015,1027,1032,1037],{"type":56,"tag":82,"props":1006,"children":1007},{},[1008],{"type":62,"value":1009},"Static validation of generated source.",{"type":56,"tag":82,"props":1011,"children":1012},{},[1013],{"type":62,"value":1014},"Docker image build.",{"type":56,"tag":82,"props":1016,"children":1017},{},[1018,1020,1026],{"type":62,"value":1019},"Service launch with ",{"type":56,"tag":171,"props":1021,"children":1023},{"className":1022},[],[1024],{"type":62,"value":1025},"deploy\u002F.env",{"type":62,"value":939},{"type":56,"tag":82,"props":1028,"children":1029},{},[1030],{"type":62,"value":1031},"API unit tests.",{"type":56,"tag":82,"props":1033,"children":1034},{},[1035],{"type":62,"value":1036},"File-video API evaluation.",{"type":56,"tag":82,"props":1038,"children":1039},{},[1040],{"type":62,"value":1041},"Latency, camera, or Kafka evaluation as applicable.",{"type":56,"tag":65,"props":1043,"children":1044},{},[1045,1050],{"type":56,"tag":363,"props":1046,"children":1047},{},[1048],{"type":62,"value":1049},"Fix any issues found during evaluation before proceeding.",{"type":62,"value":1051}," Record commands, results, timings, and fixes in the evaluation report.",{"type":56,"tag":415,"props":1053,"children":1054},{},[],{"type":56,"tag":419,"props":1056,"children":1058},{"id":1057},"phase-3-build-vss-sop-blueprint-vss-sop-build",[1059],{"type":62,"value":1060},"Phase 3 — Build VSS-SOP Blueprint (vss-sop-build)",{"type":56,"tag":65,"props":1062,"children":1063},{},[1064,1066,1070],{"type":62,"value":1065},"Run the ",{"type":56,"tag":363,"props":1067,"children":1068},{},[1069],{"type":62,"value":367},{"type":62,"value":1071}," skill from the blueprint repo root. This scaffolds the SOP blueprint layer on top of VSS 3.1:",{"type":56,"tag":438,"props":1073,"children":1075},{"className":440,"code":1074,"language":442,"meta":443,"style":443},"cd \u003Cbp-repo>\n# Follow vss-sop-build SKILL.md Quick Start steps 1–10\n",[1076],{"type":56,"tag":171,"props":1077,"children":1078},{"__ignoreMap":443},[1079,1102],{"type":56,"tag":449,"props":1080,"children":1081},{"class":451,"line":452},[1082,1086,1090,1094,1098],{"type":56,"tag":449,"props":1083,"children":1084},{"style":456},[1085],{"type":62,"value":459},{"type":56,"tag":449,"props":1087,"children":1088},{"style":462},[1089],{"type":62,"value":465},{"type":56,"tag":449,"props":1091,"children":1092},{"style":468},[1093],{"type":62,"value":471},{"type":56,"tag":449,"props":1095,"children":1096},{"style":474},[1097],{"type":62,"value":477},{"type":56,"tag":449,"props":1099,"children":1100},{"style":462},[1101],{"type":62,"value":482},{"type":56,"tag":449,"props":1103,"children":1104},{"class":451,"line":485},[1105],{"type":56,"tag":449,"props":1106,"children":1107},{"style":489},[1108],{"type":62,"value":1109},"# Follow vss-sop-build SKILL.md Quick Start steps 1–10\n",{"type":56,"tag":65,"props":1111,"children":1112},{},[1113],{"type":62,"value":1114},"Key stages:",{"type":56,"tag":667,"props":1116,"children":1117},{},[1118,1123,1128,1133,1138,1143,1148,1167],{"type":56,"tag":82,"props":1119,"children":1120},{},[1121],{"type":62,"value":1122},"Clone upstream VSS 3.1.0 repository.",{"type":56,"tag":82,"props":1124,"children":1125},{},[1126],{"type":62,"value":1127},"Copy all reference files verbatim.",{"type":56,"tag":82,"props":1129,"children":1130},{},[1131],{"type":62,"value":1132},"Copy and modify foundational services for SOP.",{"type":56,"tag":82,"props":1134,"children":1135},{},[1136],{"type":62,"value":1137},"Copy and modify VIOS for SOP.",{"type":56,"tag":82,"props":1139,"children":1140},{},[1141],{"type":62,"value":1142},"Copy agents from upstream and apply SOP profile.",{"type":56,"tag":82,"props":1144,"children":1145},{},[1146],{"type":62,"value":1147},"Copy NIM from upstream with SOP conventions.",{"type":56,"tag":82,"props":1149,"children":1150},{},[1151,1153,1159,1161,1166],{"type":62,"value":1152},"Build ",{"type":56,"tag":171,"props":1154,"children":1156},{"className":1155},[],[1157],{"type":62,"value":1158},"ds-sop:1.0.0",{"type":62,"value":1160}," image from the generated ",{"type":56,"tag":171,"props":1162,"children":1164},{"className":1163},[],[1165],{"type":62,"value":767},{"type":62,"value":939},{"type":56,"tag":82,"props":1168,"children":1169},{},[1170,1172,1178],{"type":62,"value":1171},"Verify the build with ",{"type":56,"tag":171,"props":1173,"children":1175},{"className":1174},[],[1176],{"type":62,"value":1177},"verify_build.sh",{"type":62,"value":939},{"type":56,"tag":65,"props":1180,"children":1181},{},[1182,1184,1191],{"type":62,"value":1183},"See ",{"type":56,"tag":1185,"props":1186,"children":1188},"a",{"href":1187},"..\u002Fvss-sop-build\u002FSKILL.md",[1189],{"type":62,"value":1190},"vss-sop-build SKILL.md",{"type":62,"value":1192}," for full details.",{"type":56,"tag":415,"props":1194,"children":1195},{},[],{"type":56,"tag":419,"props":1197,"children":1199},{"id":1198},"phase-4-deploy-vss-sop-blueprint-vss-sop-deploy",[1200],{"type":62,"value":1201},"Phase 4 — Deploy VSS-SOP Blueprint (vss-sop-deploy)",{"type":56,"tag":65,"props":1203,"children":1204},{},[1205,1206,1210],{"type":62,"value":1065},{"type":56,"tag":363,"props":1207,"children":1208},{},[1209],{"type":62,"value":390},{"type":62,"value":1211}," skill from the blueprint repo root:",{"type":56,"tag":438,"props":1213,"children":1215},{"className":440,"code":1214,"language":442,"meta":443,"style":443},"cd \u003Cbp-repo>\n# Follow vss-sop-deploy SKILL.md Phases 0–5\n",[1216],{"type":56,"tag":171,"props":1217,"children":1218},{"__ignoreMap":443},[1219,1242],{"type":56,"tag":449,"props":1220,"children":1221},{"class":451,"line":452},[1222,1226,1230,1234,1238],{"type":56,"tag":449,"props":1223,"children":1224},{"style":456},[1225],{"type":62,"value":459},{"type":56,"tag":449,"props":1227,"children":1228},{"style":462},[1229],{"type":62,"value":465},{"type":56,"tag":449,"props":1231,"children":1232},{"style":468},[1233],{"type":62,"value":471},{"type":56,"tag":449,"props":1235,"children":1236},{"style":474},[1237],{"type":62,"value":477},{"type":56,"tag":449,"props":1239,"children":1240},{"style":462},[1241],{"type":62,"value":482},{"type":56,"tag":449,"props":1243,"children":1244},{"class":451,"line":485},[1245],{"type":56,"tag":449,"props":1246,"children":1247},{"style":489},[1248],{"type":62,"value":1249},"# Follow vss-sop-deploy SKILL.md Phases 0–5\n",{"type":56,"tag":65,"props":1251,"children":1252},{},[1253],{"type":62,"value":1254},"Deployment phases:",{"type":56,"tag":78,"props":1256,"children":1257},{},[1258,1283,1293,1311,1321,1345],{"type":56,"tag":82,"props":1259,"children":1260},{},[1261,1266,1268,1274,1276,1281],{"type":56,"tag":363,"props":1262,"children":1263},{},[1264],{"type":62,"value":1265},"Phase 0:",{"type":62,"value":1267}," Prerequisites check and auto-install (",{"type":56,"tag":171,"props":1269,"children":1271},{"className":1270},[],[1272],{"type":62,"value":1273},"preflight_check.sh --bp-repo . --fix",{"type":62,"value":1275},") — already run in ",{"type":56,"tag":363,"props":1277,"children":1278},{},[1279],{"type":62,"value":1280},"Phase Pre",{"type":62,"value":1282}," above; re-runs as an idempotent safety check.",{"type":56,"tag":82,"props":1284,"children":1285},{},[1286,1291],{"type":56,"tag":363,"props":1287,"children":1288},{},[1289],{"type":62,"value":1290},"Phase 1:",{"type":62,"value":1292}," Download models & assets (already done in Phase 0 above — will skip if present).",{"type":56,"tag":82,"props":1294,"children":1295},{},[1296,1301,1303,1309],{"type":56,"tag":363,"props":1297,"children":1298},{},[1299],{"type":62,"value":1300},"Phase 2:",{"type":62,"value":1302}," Start RTSP server simulation (",{"type":56,"tag":171,"props":1304,"children":1306},{"className":1305},[],[1307],{"type":62,"value":1308},"start_rtsp_server.sh --bp-repo .",{"type":62,"value":1310},").",{"type":56,"tag":82,"props":1312,"children":1313},{},[1314,1319],{"type":56,"tag":363,"props":1315,"children":1316},{},[1317],{"type":62,"value":1318},"Phase 3:",{"type":62,"value":1320}," Build DS-SOP Docker image & verify components (already built in Phase 3 above — verify only).",{"type":56,"tag":82,"props":1322,"children":1323},{},[1324,1329,1331,1337,1338,1344],{"type":56,"tag":363,"props":1325,"children":1326},{},[1327],{"type":62,"value":1328},"Phase 4:",{"type":62,"value":1330}," Configure & deploy the blueprint (",{"type":56,"tag":171,"props":1332,"children":1334},{"className":1333},[],[1335],{"type":62,"value":1336},"configure_blueprint.sh",{"type":62,"value":871},{"type":56,"tag":171,"props":1339,"children":1341},{"className":1340},[],[1342],{"type":62,"value":1343},"deploy.sh",{"type":62,"value":1310},{"type":56,"tag":82,"props":1346,"children":1347},{},[1348,1353,1355,1361],{"type":56,"tag":363,"props":1349,"children":1350},{},[1351],{"type":62,"value":1352},"Phase 5:",{"type":62,"value":1354}," Test with RTSP stream (",{"type":56,"tag":171,"props":1356,"children":1358},{"className":1357},[],[1359],{"type":62,"value":1360},"test_rtsp.sh --bp-repo .",{"type":62,"value":1310},{"type":56,"tag":65,"props":1363,"children":1364},{},[1365,1366,1372],{"type":62,"value":1183},{"type":56,"tag":1185,"props":1367,"children":1369},{"href":1368},"..\u002Fvss-sop-deploy\u002FSKILL.md",[1370],{"type":62,"value":1371},"vss-sop-deploy SKILL.md",{"type":62,"value":1192},{"type":56,"tag":415,"props":1374,"children":1375},{},[],{"type":56,"tag":419,"props":1377,"children":1379},{"id":1378},"phase-5-test-vss-sop-blueprint-vss-sop-test",[1380],{"type":62,"value":1381},"Phase 5 — Test VSS-SOP Blueprint (vss-sop-test)",{"type":56,"tag":65,"props":1383,"children":1384},{},[1385,1386,1390],{"type":62,"value":1065},{"type":56,"tag":363,"props":1387,"children":1388},{},[1389],{"type":62,"value":412},{"type":62,"value":1211},{"type":56,"tag":438,"props":1392,"children":1394},{"className":440,"code":1393,"language":442,"meta":443,"style":443},"cd \u003Cbp-repo>\npython agentic\u002Fvss-sop-skills\u002Fvss-sop-test\u002Fscripts\u002Fvss_sop_test.py \\\n  --bp-repo . \\\n  --env-file deployments\u002Fsop\u002F.env\n",[1395],{"type":56,"tag":171,"props":1396,"children":1397},{"__ignoreMap":443},[1398,1421,1439,1455],{"type":56,"tag":449,"props":1399,"children":1400},{"class":451,"line":452},[1401,1405,1409,1413,1417],{"type":56,"tag":449,"props":1402,"children":1403},{"style":456},[1404],{"type":62,"value":459},{"type":56,"tag":449,"props":1406,"children":1407},{"style":462},[1408],{"type":62,"value":465},{"type":56,"tag":449,"props":1410,"children":1411},{"style":468},[1412],{"type":62,"value":471},{"type":56,"tag":449,"props":1414,"children":1415},{"style":474},[1416],{"type":62,"value":477},{"type":56,"tag":449,"props":1418,"children":1419},{"style":462},[1420],{"type":62,"value":482},{"type":56,"tag":449,"props":1422,"children":1423},{"class":451,"line":485},[1424,1429,1434],{"type":56,"tag":449,"props":1425,"children":1426},{"style":499},[1427],{"type":62,"value":1428},"python",{"type":56,"tag":449,"props":1430,"children":1431},{"style":468},[1432],{"type":62,"value":1433}," agentic\u002Fvss-sop-skills\u002Fvss-sop-test\u002Fscripts\u002Fvss_sop_test.py",{"type":56,"tag":449,"props":1435,"children":1436},{"style":474},[1437],{"type":62,"value":1438}," \\\n",{"type":56,"tag":449,"props":1440,"children":1441},{"class":451,"line":495},[1442,1447,1451],{"type":56,"tag":449,"props":1443,"children":1444},{"style":468},[1445],{"type":62,"value":1446},"  --bp-repo",{"type":56,"tag":449,"props":1448,"children":1449},{"style":468},[1450],{"type":62,"value":512},{"type":56,"tag":449,"props":1452,"children":1453},{"style":474},[1454],{"type":62,"value":1438},{"type":56,"tag":449,"props":1456,"children":1458},{"class":451,"line":1457},4,[1459,1464],{"type":56,"tag":449,"props":1460,"children":1461},{"style":468},[1462],{"type":62,"value":1463},"  --env-file",{"type":56,"tag":449,"props":1465,"children":1466},{"style":468},[1467],{"type":62,"value":1468}," deployments\u002Fsop\u002F.env\n",{"type":56,"tag":65,"props":1470,"children":1471},{},[1472],{"type":62,"value":1473},"Test phases:",{"type":56,"tag":78,"props":1475,"children":1476},{},[1477,1486,1495,1504],{"type":56,"tag":82,"props":1478,"children":1479},{},[1480,1484],{"type":56,"tag":363,"props":1481,"children":1482},{},[1483],{"type":62,"value":1290},{"type":62,"value":1485}," Service health — all required containers running.",{"type":56,"tag":82,"props":1487,"children":1488},{},[1489,1493],{"type":56,"tag":363,"props":1490,"children":1491},{},[1492],{"type":62,"value":1300},{"type":62,"value":1494}," ELK data pipeline — Elasticsearch, indices, VLM messages, Kibana fields.",{"type":56,"tag":82,"props":1496,"children":1497},{},[1498,1502],{"type":56,"tag":363,"props":1499,"children":1500},{},[1501],{"type":62,"value":1318},{"type":62,"value":1503}," VIOS recording & livestream — sensors, streams, recording, livestream.",{"type":56,"tag":82,"props":1505,"children":1506},{},[1507,1511],{"type":56,"tag":363,"props":1508,"children":1509},{},[1510],{"type":62,"value":1328},{"type":62,"value":1512}," VSS Agent end-to-end — MCP, LLM, VLM, snapshot, video, report generation.",{"type":56,"tag":65,"props":1514,"children":1515},{},[1516,1517,1523],{"type":62,"value":1183},{"type":56,"tag":1185,"props":1518,"children":1520},{"href":1519},"..\u002Fvss-sop-test\u002FSKILL.md",[1521],{"type":62,"value":1522},"vss-sop-test SKILL.md",{"type":62,"value":1192},{"type":56,"tag":415,"props":1525,"children":1526},{},[],{"type":56,"tag":71,"props":1528,"children":1530},{"id":1529},"error-handling",[1531],{"type":62,"value":1532},"Error Handling",{"type":56,"tag":419,"props":1534,"children":1536},{"id":1535},"pass-criteria-and-verification",[1537],{"type":62,"value":1538},"Pass Criteria and Verification",{"type":56,"tag":65,"props":1540,"children":1541},{},[1542],{"type":62,"value":1543},"Ensure all unit tests, build checks, deploy checks, and test phases pass with zero failures.",{"type":56,"tag":78,"props":1545,"children":1546},{},[1547,1552,1557,1562,1567],{"type":56,"tag":82,"props":1548,"children":1549},{},[1550],{"type":62,"value":1551},"Do not classify any failure as environmental, intermittent, infrastructure, or acceptable. Treat every failure as a bug to be fixed.",{"type":56,"tag":82,"props":1553,"children":1554},{},[1555],{"type":62,"value":1556},"Do not proceed to the next phase if the current phase has any failing test. Fix all failures first.",{"type":56,"tag":82,"props":1558,"children":1559},{},[1560],{"type":62,"value":1561},"Do not declare completion if any test is failing. A result like \"36\u002F38 checks pass\" is a failure, not a success.",{"type":56,"tag":82,"props":1563,"children":1564},{},[1565],{"type":62,"value":1566},"If a test times out, fix the timeout or the underlying latency rather than dismissing it.",{"type":56,"tag":82,"props":1568,"children":1569},{},[1570],{"type":62,"value":1571},"If stuck on the same failure after three fix attempts, stop and ask the user for help. Do not skip the failure.",{"type":56,"tag":419,"props":1573,"children":1575},{"id":1574},"auto-debug-and-fix-loop",[1576],{"type":62,"value":1577},"Auto-Debug and Fix Loop",{"type":56,"tag":65,"props":1579,"children":1580},{},[1581],{"type":62,"value":1582},"When any phase fails:",{"type":56,"tag":667,"props":1584,"children":1585},{},[1586,1591,1603,1608,1613],{"type":56,"tag":82,"props":1587,"children":1588},{},[1589],{"type":62,"value":1590},"Read the failure output and auto-debug hints.",{"type":56,"tag":82,"props":1592,"children":1593},{},[1594,1596,1602],{"type":62,"value":1595},"Collect container logs (",{"type":56,"tag":171,"props":1597,"children":1599},{"className":1598},[],[1600],{"type":62,"value":1601},"docker logs \u003Ccontainer> --tail 50",{"type":62,"value":1310},{"type":56,"tag":82,"props":1604,"children":1605},{},[1606],{"type":62,"value":1607},"Apply the fix (env, compose, patches, image rebuild, source regeneration).",{"type":56,"tag":82,"props":1609,"children":1610},{},[1611],{"type":62,"value":1612},"Re-run the failing phase to confirm the fix.",{"type":56,"tag":82,"props":1614,"children":1615},{},[1616],{"type":62,"value":1617},"Repeat until all checks in that phase pass.",{"type":56,"tag":65,"props":1619,"children":1620},{},[1621,1623,1628,1630,1635],{"type":62,"value":1622},"If the root cause is in the generated DS-SOP source code, update ",{"type":56,"tag":171,"props":1624,"children":1626},{"className":1625},[],[1627],{"type":62,"value":205},{"type":62,"value":1629}," references and regenerate. If the root cause is in the build\u002Fdeploy config, update ",{"type":56,"tag":171,"props":1631,"children":1633},{"className":1632},[],[1634],{"type":62,"value":367},{"type":62,"value":1636}," references.",{"type":56,"tag":415,"props":1638,"children":1639},{},[],{"type":56,"tag":71,"props":1641,"children":1643},{"id":1642},"examples",[1644],{"type":62,"value":1645},"Examples",{"type":56,"tag":65,"props":1647,"children":1648},{},[1649],{"type":62,"value":1650},"After all phases complete with 100% pass rate, present a summary table:",{"type":56,"tag":104,"props":1652,"children":1653},{},[1654,1675],{"type":56,"tag":108,"props":1655,"children":1656},{},[1657],{"type":56,"tag":112,"props":1658,"children":1659},{},[1660,1665,1670],{"type":56,"tag":116,"props":1661,"children":1662},{},[1663],{"type":62,"value":1664},"Category",{"type":56,"tag":116,"props":1666,"children":1667},{},[1668],{"type":62,"value":1669},"Status",{"type":56,"tag":116,"props":1671,"children":1672},{},[1673],{"type":62,"value":1674},"Details",{"type":56,"tag":127,"props":1676,"children":1677},{},[1678,1696,1713,1730,1747,1764,1781],{"type":56,"tag":112,"props":1679,"children":1680},{},[1681,1686,1691],{"type":56,"tag":134,"props":1682,"children":1683},{},[1684],{"type":62,"value":1685},"Phase Pre: Prerequisites",{"type":56,"tag":134,"props":1687,"children":1688},{},[1689],{"type":62,"value":1690},"Pass\u002FFail",{"type":56,"tag":134,"props":1692,"children":1693},{},[1694],{"type":62,"value":1695},"Driver, Docker, Toolkit, NGC config",{"type":56,"tag":112,"props":1697,"children":1698},{},[1699,1704,1708],{"type":56,"tag":134,"props":1700,"children":1701},{},[1702],{"type":62,"value":1703},"Phase 0: Assets",{"type":56,"tag":134,"props":1705,"children":1706},{},[1707],{"type":62,"value":1690},{"type":56,"tag":134,"props":1709,"children":1710},{},[1711],{"type":62,"value":1712},"Models and configs at \u002Fopt",{"type":56,"tag":112,"props":1714,"children":1715},{},[1716,1721,1725],{"type":56,"tag":134,"props":1717,"children":1718},{},[1719],{"type":62,"value":1720},"Phase 1: Generate",{"type":56,"tag":134,"props":1722,"children":1723},{},[1724],{"type":62,"value":1690},{"type":56,"tag":134,"props":1726,"children":1727},{},[1728],{"type":62,"value":1729},"..\u002Fds_sop_microservice with RTSP",{"type":56,"tag":112,"props":1731,"children":1732},{},[1733,1738,1742],{"type":56,"tag":134,"props":1734,"children":1735},{},[1736],{"type":62,"value":1737},"Phase 2: Evaluate",{"type":56,"tag":134,"props":1739,"children":1740},{},[1741],{"type":62,"value":1690},{"type":56,"tag":134,"props":1743,"children":1744},{},[1745],{"type":62,"value":1746},"Unit tests + API evaluation",{"type":56,"tag":112,"props":1748,"children":1749},{},[1750,1755,1759],{"type":56,"tag":134,"props":1751,"children":1752},{},[1753],{"type":62,"value":1754},"Phase 3: Build",{"type":56,"tag":134,"props":1756,"children":1757},{},[1758],{"type":62,"value":1690},{"type":56,"tag":134,"props":1760,"children":1761},{},[1762],{"type":62,"value":1763},"Per-stage build status",{"type":56,"tag":112,"props":1765,"children":1766},{},[1767,1772,1776],{"type":56,"tag":134,"props":1768,"children":1769},{},[1770],{"type":62,"value":1771},"Phase 4: Deploy",{"type":56,"tag":134,"props":1773,"children":1774},{},[1775],{"type":62,"value":1690},{"type":56,"tag":134,"props":1777,"children":1778},{},[1779],{"type":62,"value":1780},"Per-phase deploy status",{"type":56,"tag":112,"props":1782,"children":1783},{},[1784,1789,1793],{"type":56,"tag":134,"props":1785,"children":1786},{},[1787],{"type":62,"value":1788},"Phase 5: Test",{"type":56,"tag":134,"props":1790,"children":1791},{},[1792],{"type":62,"value":1690},{"type":56,"tag":134,"props":1794,"children":1795},{},[1796],{"type":62,"value":1797},"Per-phase test status (must be 100%)",{"type":56,"tag":65,"props":1799,"children":1800},{},[1801],{"type":62,"value":1802},"Include:",{"type":56,"tag":78,"props":1804,"children":1805},{},[1806,1825],{"type":56,"tag":82,"props":1807,"children":1808},{},[1809,1811,1816,1818,1823],{"type":62,"value":1810},"Any ",{"type":56,"tag":171,"props":1812,"children":1814},{"className":1813},[],[1815],{"type":62,"value":367},{"type":62,"value":1817}," or ",{"type":56,"tag":171,"props":1819,"children":1821},{"className":1820},[],[1822],{"type":62,"value":205},{"type":62,"value":1824}," files updated during debugging.",{"type":56,"tag":82,"props":1826,"children":1827},{},[1828],{"type":62,"value":1829},"Service endpoints: VSS-UI, Kibana, VIOS-UI, Grafana.",{"type":56,"tag":65,"props":1831,"children":1832},{},[1833,1835,1840],{"type":62,"value":1834},"If any test is still failing, the report must clearly state ",{"type":56,"tag":363,"props":1836,"children":1837},{},[1838],{"type":62,"value":1839},"\"Pipeline Incomplete — X failures remain\"",{"type":62,"value":1841}," and list them. Do not present a partial success as completion.",{"type":56,"tag":419,"props":1843,"children":1845},{"id":1844},"service-endpoints-after-successful-deploy",[1846],{"type":62,"value":1847},"Service Endpoints (after successful deploy)",{"type":56,"tag":104,"props":1849,"children":1850},{},[1851,1867],{"type":56,"tag":108,"props":1852,"children":1853},{},[1854],{"type":56,"tag":112,"props":1855,"children":1856},{},[1857,1862],{"type":56,"tag":116,"props":1858,"children":1859},{},[1860],{"type":62,"value":1861},"Service",{"type":56,"tag":116,"props":1863,"children":1864},{},[1865],{"type":62,"value":1866},"Endpoint",{"type":56,"tag":127,"props":1868,"children":1869},{},[1870,1887,1904,1921,1938],{"type":56,"tag":112,"props":1871,"children":1872},{},[1873,1878],{"type":56,"tag":134,"props":1874,"children":1875},{},[1876],{"type":62,"value":1877},"VSS-UI",{"type":56,"tag":134,"props":1879,"children":1880},{},[1881],{"type":56,"tag":171,"props":1882,"children":1884},{"className":1883},[],[1885],{"type":62,"value":1886},"http:\u002F\u002F\u003CEXTERNAL_IP>:3000",{"type":56,"tag":112,"props":1888,"children":1889},{},[1890,1895],{"type":56,"tag":134,"props":1891,"children":1892},{},[1893],{"type":62,"value":1894},"Kibana",{"type":56,"tag":134,"props":1896,"children":1897},{},[1898],{"type":56,"tag":171,"props":1899,"children":1901},{"className":1900},[],[1902],{"type":62,"value":1903},"http:\u002F\u002F\u003CEXTERNAL_IP>:5601\u002Fapp\u002Fhome#\u002F",{"type":56,"tag":112,"props":1905,"children":1906},{},[1907,1912],{"type":56,"tag":134,"props":1908,"children":1909},{},[1910],{"type":62,"value":1911},"VIOS-UI",{"type":56,"tag":134,"props":1913,"children":1914},{},[1915],{"type":56,"tag":171,"props":1916,"children":1918},{"className":1917},[],[1919],{"type":62,"value":1920},"http:\u002F\u002F\u003CEXTERNAL_IP>:30888\u002Fvst\u002F#\u002Fdashboard",{"type":56,"tag":112,"props":1922,"children":1923},{},[1924,1929],{"type":56,"tag":134,"props":1925,"children":1926},{},[1927],{"type":62,"value":1928},"Grafana",{"type":56,"tag":134,"props":1930,"children":1931},{},[1932],{"type":56,"tag":171,"props":1933,"children":1935},{"className":1934},[],[1936],{"type":62,"value":1937},"http:\u002F\u002F\u003CEXTERNAL_IP>:35000\u002F",{"type":56,"tag":112,"props":1939,"children":1940},{},[1941,1946],{"type":56,"tag":134,"props":1942,"children":1943},{},[1944],{"type":62,"value":1945},"Phoenix (Telemetry)",{"type":56,"tag":134,"props":1947,"children":1948},{},[1949],{"type":56,"tag":171,"props":1950,"children":1952},{"className":1951},[],[1953],{"type":62,"value":1954},"http:\u002F\u002F\u003CEXTERNAL_IP>:6006\u002Fprojects",{"type":56,"tag":415,"props":1956,"children":1957},{},[],{"type":56,"tag":71,"props":1959,"children":1961},{"id":1960},"license",[1962],{"type":62,"value":1963},"License",{"type":56,"tag":65,"props":1965,"children":1966},{},[1967,1969,1977,1979,1986],{"type":62,"value":1968},"Use of this skill is governed by the ",{"type":56,"tag":1185,"props":1970,"children":1974},{"href":1971,"rel":1972},"https:\u002F\u002Fcreativecommons.org\u002Flicenses\u002Fby\u002F4.0\u002Flegalcode.en",[1973],"nofollow",[1975],{"type":62,"value":1976},"Creative Commons Attribution 4.0 International License (CC BY 4.0)",{"type":62,"value":1978}," and the ",{"type":56,"tag":1185,"props":1980,"children":1983},{"href":1981,"rel":1982},"https:\u002F\u002Fwww.apache.org\u002Flicenses\u002FLICENSE-2.0",[1973],[1984],{"type":62,"value":1985},"Apache License, Version 2.0",{"type":62,"value":939},{"type":56,"tag":1988,"props":1989,"children":1990},"style",{},[1991],{"type":62,"value":1992},"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":1994,"total":2085},[1995,2003,2018,2031,2049,2061,2072],{"slug":4,"name":4,"fn":5,"description":6,"org":1996,"tags":1997,"stars":26,"repoUrl":27,"updatedAt":28},{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1998,1999,2000,2001,2002],{"name":17,"slug":18,"type":15},{"name":21,"slug":22,"type":15},{"name":13,"slug":14,"type":15},{"name":9,"slug":8,"type":15},{"name":24,"slug":25,"type":15},{"slug":2004,"name":2004,"fn":2005,"description":2006,"org":2007,"tags":2008,"stars":26,"repoUrl":27,"updatedAt":2017},"sop-by-action-eval","run by-action VLM evaluations","Use when running by-action VLM evaluation (per-action-clip inference + accuracy metrics) against the BP evaluation-ms HTTP API. Invoked as \u002Fsop-by-action-eval \u003Cinputs.yaml> [natural language parameter overrides]",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[2009,2012,2015,2016],{"name":2010,"slug":2011,"type":15},"Evals","evals",{"name":2013,"slug":2014,"type":15},"Machine Learning","machine-learning",{"name":13,"slug":14,"type":15},{"name":9,"slug":8,"type":15},"2026-07-14T05:36:13.260585",{"slug":2019,"name":2019,"fn":2020,"description":2021,"org":2022,"tags":2023,"stars":26,"repoUrl":27,"updatedAt":2030},"sop-cr-finetuning","fine-tune VLM models for SOP monitoring","Fine-tune Cosmos-Reason2 (CR2) VLM for SOP monitoring. Use when you need to launch and monitor a VLM training run with a given dataset ID.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[2024,2027,2028,2029],{"name":2025,"slug":2026,"type":15},"AI Infrastructure","ai-infrastructure",{"name":2013,"slug":2014,"type":15},{"name":13,"slug":14,"type":15},{"name":9,"slug":8,"type":15},"2026-07-14T05:36:22.874218",{"slug":2032,"name":2032,"fn":2033,"description":2034,"org":2035,"tags":2036,"stars":26,"repoUrl":27,"updatedAt":2048},"sop-data-augmentation","augment annotated datasets","Use when the user wants to run data augmentation on an annotated dataset, configure augmentation parameters, check augmentation status, or understand what each QA augmentation type does (BCQ, MCQ, GQA, DMCQ, DSQA, ENQA)",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[2037,2040,2043,2044,2045],{"name":2038,"slug":2039,"type":15},"Data Engineering","data-engineering",{"name":2041,"slug":2042,"type":15},"Datasets","datasets",{"name":2013,"slug":2014,"type":15},{"name":9,"slug":8,"type":15},{"name":2046,"slug":2047,"type":15},"QA","qa","2026-07-14T05:36:20.305657",{"slug":2050,"name":2050,"fn":2051,"description":2052,"org":2053,"tags":2054,"stars":26,"repoUrl":27,"updatedAt":2060},"sop-ddm-finetuning","fine-tune DDM-Net models for SOP monitoring","Fine-tune DDM-Net temporal boundary detector for SOP monitoring. Use when you need to launch and monitor a DDM-Net training run with a given dataset ID.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[2055,2058,2059],{"name":2056,"slug":2057,"type":15},"Deep Learning","deep-learning",{"name":13,"slug":14,"type":15},{"name":9,"slug":8,"type":15},"2026-07-14T05:35:16.203684",{"slug":2062,"name":2062,"fn":2063,"description":2064,"org":2065,"tags":2066,"stars":26,"repoUrl":27,"updatedAt":2071},"sop-e2e-inference","run end-to-end inference evaluations","Use when running the e2e evaluation pipeline (temporal segmentation + action recognition + accuracy) against the BP evaluation-ms HTTP API. Invoked as \u002Fsop-e2e-inference \u003Cinputs.yaml> [natural language parameter overrides]",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[2067,2068,2069,2070],{"name":2010,"slug":2011,"type":15},{"name":2013,"slug":2014,"type":15},{"name":13,"slug":14,"type":15},{"name":9,"slug":8,"type":15},"2026-07-14T05:36:15.23958",{"slug":2073,"name":2073,"fn":2074,"description":2075,"org":2076,"tags":2077,"stars":26,"repoUrl":27,"updatedAt":2084},"sop-ft-orchestrate","orchestrate SOP fine-tuning pipelines","Autonomous end-to-end orchestrator for SOP fine-tuning. Runs the full Import → Augment → DDM Train → VLM Train → Evaluate → RCA loop. Interprets RCA findings across DDM, VLM and augment axes, applies config fixes autonomously, and iterates until success criteria are met or max_pipeline_iterations reached. Call with a path to an inputs.yaml or with natural language.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[2078,2079,2080,2081],{"name":17,"slug":18,"type":15},{"name":2013,"slug":2014,"type":15},{"name":9,"slug":8,"type":15},{"name":2082,"slug":2083,"type":15},"Orchestration","orchestration","2026-07-14T05:36:28.162686",11,{"items":2087,"total":2239},[2088,2106,2123,2134,2146,2160,2173,2185,2196,2207,2221,2230],{"slug":2089,"name":2089,"fn":2090,"description":2091,"org":2092,"tags":2093,"stars":2103,"repoUrl":2104,"updatedAt":2105},"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},[2094,2097,2100],{"name":2095,"slug":2096,"type":15},"Documentation","documentation",{"name":2098,"slug":2099,"type":15},"MCP","mcp",{"name":2101,"slug":2102,"type":15},"Search","search",21777,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FNemoClaw","2026-07-20T06:00:01.461044",{"slug":2107,"name":2107,"fn":2108,"description":2109,"org":2110,"tags":2111,"stars":2120,"repoUrl":2121,"updatedAt":2122},"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},[2112,2115,2118],{"name":2113,"slug":2114,"type":15},"Containers","containers",{"name":2116,"slug":2117,"type":15},"Deployment","deployment",{"name":2119,"slug":1428,"type":15},"Python",17049,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FMegatron-LM","2026-07-27T06:06:11.249662",{"slug":2124,"name":2124,"fn":2125,"description":2126,"org":2127,"tags":2128,"stars":2120,"repoUrl":2121,"updatedAt":2133},"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},[2129,2132],{"name":2130,"slug":2131,"type":15},"CI\u002FCD","ci-cd",{"name":2116,"slug":2117,"type":15},"2026-07-14T05:25:59.97109",{"slug":2135,"name":2135,"fn":2136,"description":2137,"org":2138,"tags":2139,"stars":2120,"repoUrl":2121,"updatedAt":2145},"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},[2140,2141,2142],{"name":2130,"slug":2131,"type":15},{"name":2116,"slug":2117,"type":15},{"name":2143,"slug":2144,"type":15},"GitHub","github","2026-07-27T06:06:12.278222",{"slug":2147,"name":2147,"fn":2148,"description":2149,"org":2150,"tags":2151,"stars":2120,"repoUrl":2121,"updatedAt":2159},"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},[2152,2155,2156],{"name":2153,"slug":2154,"type":15},"Debugging","debugging",{"name":2143,"slug":2144,"type":15},{"name":2157,"slug":2158,"type":15},"Triage","triage","2026-07-14T05:25:57.442089",{"slug":2161,"name":2161,"fn":2162,"description":2163,"org":2164,"tags":2165,"stars":2120,"repoUrl":2121,"updatedAt":2172},"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},[2166,2169],{"name":2167,"slug":2168,"type":15},"Best Practices","best-practices",{"name":2170,"slug":2171,"type":15},"Code Analysis","code-analysis","2026-07-14T05:25:56.18433",{"slug":2174,"name":2174,"fn":2175,"description":2176,"org":2177,"tags":2178,"stars":2120,"repoUrl":2121,"updatedAt":2184},"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},[2179,2180,2183],{"name":2013,"slug":2014,"type":15},{"name":2181,"slug":2182,"type":15},"Migration","migration",{"name":9,"slug":8,"type":15},"2026-07-17T06:07:11.777011",{"slug":2186,"name":2186,"fn":2187,"description":2188,"org":2189,"tags":2190,"stars":2120,"repoUrl":2121,"updatedAt":2195},"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},[2191,2192],{"name":2046,"slug":2047,"type":15},{"name":2193,"slug":2194,"type":15},"Testing","testing","2026-07-14T05:25:53.673039",{"slug":2197,"name":2197,"fn":2198,"description":2199,"org":2200,"tags":2201,"stars":2120,"repoUrl":2121,"updatedAt":2206},"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},[2202,2203],{"name":2116,"slug":2117,"type":15},{"name":2204,"slug":2205,"type":15},"Infrastructure","infrastructure","2026-07-14T05:25:49.362534",{"slug":2208,"name":2208,"fn":2209,"description":2210,"org":2211,"tags":2212,"stars":2120,"repoUrl":2121,"updatedAt":2220},"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},[2213,2216,2217],{"name":2214,"slug":2215,"type":15},"Code Review","code-review",{"name":2143,"slug":2144,"type":15},{"name":2218,"slug":2219,"type":15},"Pull Requests","pull-requests","2026-07-14T05:26:01.226578",{"slug":2222,"name":2222,"fn":2223,"description":2224,"org":2225,"tags":2226,"stars":2120,"repoUrl":2121,"updatedAt":2229},"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},[2227,2228],{"name":2046,"slug":2047,"type":15},{"name":2193,"slug":2194,"type":15},"2026-07-14T05:25:54.928983",{"slug":2231,"name":2231,"fn":2232,"description":2233,"org":2234,"tags":2235,"stars":2120,"repoUrl":2121,"updatedAt":2238},"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},[2236,2237],{"name":17,"slug":18,"type":15},{"name":2130,"slug":2131,"type":15},"2026-07-30T05:29:03.275638",496]