[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-nvidia-i4h-workflow-e2e":3,"mdc-s2g2b5-key":34,"related-repo-nvidia-i4h-workflow-e2e":1548,"related-org-nvidia-i4h-workflow-e2e":1652},{"slug":4,"name":4,"fn":5,"description":6,"org":7,"tags":11,"stars":23,"repoUrl":24,"updatedAt":25,"license":26,"forks":27,"topics":28,"repo":29,"sourceUrl":32,"mdContent":33},"i4h-workflow-e2e","run end-to-end agentic pipelines","Run the full end-to-end agentic pipeline (record → mimic → annotate → replay → convert → visualize → finetune → validate). Use when asked to run the whole pipeline or do an e2e, smoke, or demo run.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},"nvidia","NVIDIA","https:\u002F\u002Fpexgzepcugksgbtrxkhf.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Forg-logos\u002Fnvidia.png",[12,16,19,22],{"name":13,"slug":14,"type":15},"Robotics","robotics","tag",{"name":17,"slug":18,"type":15},"Automation","automation",{"name":20,"slug":21,"type":15},"Machine Learning","machine-learning",{"name":9,"slug":8,"type":15},2473,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fskills","2026-07-23T06:06:06.300182","Apache-2.0",281,[],{"repoUrl":24,"stars":23,"forks":27,"topics":30,"description":31},[],"AI agent skills published by NVIDIA","https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fskills\u002Ftree\u002FHEAD\u002Fskills\u002Fi4h-workflow-e2e","---\nname: i4h-workflow-e2e\nversion: \"0.6.0\"\ndescription: Run the full end-to-end agentic pipeline (record → mimic → annotate → replay → convert → visualize → finetune → validate). Use when asked to run the whole pipeline or do an e2e, smoke, or demo run.\nlicense: Apache-2.0\nmetadata:\n  author: \"Isaac for Healthcare Team \u003Cisaac-for-healthcare-support@nvidia.com>\"\n  tags:\n    - isaac-for-healthcare\n    - i4h\n    - agentic-workflow\n    - end-to-end\n    - pipeline\n---\n\n# i4h Workflow — End-to-End\n\n## Purpose\n\nRun the full end-to-end agentic pipeline (record, mimic, annotate\u002Ffilter, replay, convert, visualize, finetune, validate). Use when the user asks to run the full pipeline, smoke the whole workflow, demo the workflow, or do an e2e run.\n\n## Base Code\n\nThese steps drive the i4h-workflows base code (the `workflows\u002Fagentic\u002F` tree). To reuse an existing checkout, set `I4H_WORKFLOWS` to its path (no clone happens). Otherwise this resolves the current repo, or clones to `~\u002Fi4h-workflows` — pick that default without prompting. Run every command below from the resolved root:\n\n```bash\n# Resolve the i4h-workflows base code (provides workflows\u002Fagentic\u002F).\nROOT=\"${I4H_WORKFLOWS:-$(git rev-parse --show-toplevel 2>\u002Fdev\u002Fnull)}\"\nif [ ! -d \"$ROOT\u002Fworkflows\u002Fagentic\" ]; then\n  ROOT=\"${I4H_WORKFLOWS:-$HOME\u002Fi4h-workflows}\"\n  [ -d \"$ROOT\u002Fworkflows\u002Fagentic\" ] || git clone https:\u002F\u002Fgithub.com\u002Fisaac-for-healthcare\u002Fi4h-workflows \"$ROOT\"\nfi\nexport I4H_WORKFLOWS=\"$ROOT\"; cd \"$ROOT\"\n```\n\n## Basics\n\n- **Env config (source of truth):** `workflows\u002Fagentic\u002Fconfig\u002Fenvironments\u002F\u003Cenv>.yaml` — drives every stage for `\u003Cenv>` (robot, task, policy, cameras, `arena.max_timesteps`, `dataset.*` mappings).\n- Use the e2e script for full pipeline runs.\n- For per-stage work, use the corresponding dataset\u002Ffinetune\u002Fvalidate skills.\n- `assemble_trocar` is inference-only; the e2e script skips finetune and checkpoint validation for it.\n\n## Dry Run\n\n```bash\nREPO_ROOT=\"${I4H_WORKFLOWS:-$(git rev-parse --show-toplevel 2>\u002Fdev\u002Fnull)}\"; [ -d \"$REPO_ROOT\u002Fworkflows\u002Fagentic\" ] || REPO_ROOT=\"$HOME\u002Fi4h-workflows\"\n\"${REPO_ROOT}\u002Fworkflows\u002Fagentic\u002Fscripts\u002Fe2e\u002Frun.sh\" --dry-run --env \u003Cenv>\n```\n\n## Run\n\nRun the steps below in order. Each step is a separate bash call; variables persist in the local agent's tmux session.\n\nFor Claude Code `--print` or any other noninteractive runner, keep Step 2 in the foreground. This is a validation requirement: do not use Claude background tasks, async task mode, Bash background mode, `&`, `nohup`, `tmux`, `disown`, or any detached process\u002Ftask id, and do not answer that the pipeline is still running. Do not return until `run.sh` exits and you have inspected `logs\u002FSUMMARY.txt` on success, or the failing stage log on failure. Report the run dir, skipped stages, per-stage status, key artifacts, and cleanup\u002Fstop status before finishing.\n\n### Step 1 — setup\n\n```bash\nREPO_ROOT=\"${I4H_WORKFLOWS:-$(git rev-parse --show-toplevel 2>\u002Fdev\u002Fnull)}\"; [ -d \"$REPO_ROOT\u002Fworkflows\u002Fagentic\" ] || REPO_ROOT=\"$HOME\u002Fi4h-workflows\"\n```\n\n### Step 2 — e2e pipeline\n\n```bash\n\"${REPO_ROOT}\u002Fworkflows\u002Fagentic\u002Fscripts\u002Fe2e\u002Frun.sh\" --env \u003Cenv>\n```\n\n## Flags\n\n- `--skip-mimic`, `--skip-annotate`, `--skip-replay`, `--skip-viz`\n- `--from-stage \u003Cstage> --run-dir \u003Cexisting-run>` resumes from a prior run.\n- Policy record\u002Fverify stages open the sim window by default. Set `ARENA_HEADLESS=1` before `run.sh` only when the user explicitly asks for headless\u002Fno-window execution.\n\nStages: `setup record mimic annotate replay convert viz finetune validate summary`.\n\n## Outputs\n\nThe script prints `RUN_DIR` and symlinks it to `runs\u002F.latest`. Subdirs:\n\n- `logs\u002F` — per-stage logs, `workflow.log` (full teed output), and `logs\u002FSUMMARY.txt` (the final summary report)\n- `data\u002F`\n- `lerobot\u002F`\n- `checkpoint\u002F` (trainable envs only)\n\n## Monitor\n\n`run.sh` runs every stage in the foreground and returns only when the whole pipeline ends, so track a long run from a **separate shell** (do not expect to query it from the shell that launched it):\n\n```bash\ntail -f \"${REPO_ROOT}\u002Fworkflows\u002Fagentic\u002Fruns\u002F.latest\u002Flogs\u002Fworkflow.log\"   # live per-stage progress\ncat    \"${REPO_ROOT}\u002Fworkflows\u002Fagentic\u002Fruns\u002F.latest\u002Flogs\u002FSUMMARY.txt\"     # final report (once DONE)\n```\n\n## Stop\n\n### Step 3 — stop (if needed)\n\n```bash\n\"${REPO_ROOT}\u002Fworkflows\u002Fagentic\u002Fstop.sh\" all --env \u003Cenv>\n```\n\n## Prerequisites\n\n- Workflow set up via [[i4h-workflow-setup]] (the `.venv` must exist); `setup` is also the first pipeline stage.\n- A valid `--env` name to drive the run.\n- For per-stage work, use the corresponding dataset\u002Ffinetune\u002Fvalidate skills instead.\n\n## Limitations\n\n- `assemble_trocar` is inference-only; the e2e script skips finetune and checkpoint validation for it.\n- `checkpoint\u002F` outputs are produced for trainable envs only.\n- Resuming requires both `--from-stage \u003Cstage>` and `--run-dir \u003Cexisting-run>`.\n\n## Troubleshooting\n\n- **Error:** `.venv` not found \u002F module import fails - Cause: workflow not set up. Fix: run [[i4h-workflow-setup]] first.\n- **Error:** env not recognized - Cause: wrong `--env` name. Fix: pass a valid env name; dry-run first with `--dry-run --env \u003Cenv>`.\n- **Error:** resume fails to find prior outputs - Cause: `--from-stage` used without a matching `--run-dir`. Fix: pass `--from-stage \u003Cstage> --run-dir \u003Cexisting-run>`.\n- **Error:** stale processes block a rerun - Cause: a previous pipeline session is still running. Fix: run `stop.sh all --env \u003Cenv>` before retrying.\n\n## Final Response\n\nReport env, run dir, skipped stages, per-stage success\u002Ffailure, key artifact paths.\n",{"data":35,"body":45},{"name":4,"version":36,"description":6,"license":26,"metadata":37},"0.6.0",{"author":38,"tags":39},"Isaac for Healthcare Team \u003Cisaac-for-healthcare-support@nvidia.com>",[40,41,42,43,44],"isaac-for-healthcare","i4h","agentic-workflow","end-to-end","pipeline",{"type":46,"children":47},"root",[48,57,64,70,76,106,448,454,523,529,728,734,739,797,804,942,948,998,1004,1068,1081,1087,1108,1166,1172,1189,1271,1277,1283,1339,1345,1387,1393,1435,1441,1531,1537,1542],{"type":49,"tag":50,"props":51,"children":53},"element","h1",{"id":52},"i4h-workflow-end-to-end",[54],{"type":55,"value":56},"text","i4h Workflow — End-to-End",{"type":49,"tag":58,"props":59,"children":61},"h2",{"id":60},"purpose",[62],{"type":55,"value":63},"Purpose",{"type":49,"tag":65,"props":66,"children":67},"p",{},[68],{"type":55,"value":69},"Run the full end-to-end agentic pipeline (record, mimic, annotate\u002Ffilter, replay, convert, visualize, finetune, validate). Use when the user asks to run the full pipeline, smoke the whole workflow, demo the workflow, or do an e2e run.",{"type":49,"tag":58,"props":71,"children":73},{"id":72},"base-code",[74],{"type":55,"value":75},"Base Code",{"type":49,"tag":65,"props":77,"children":78},{},[79,81,88,90,96,98,104],{"type":55,"value":80},"These steps drive the i4h-workflows base code (the ",{"type":49,"tag":82,"props":83,"children":85},"code",{"className":84},[],[86],{"type":55,"value":87},"workflows\u002Fagentic\u002F",{"type":55,"value":89}," tree). To reuse an existing checkout, set ",{"type":49,"tag":82,"props":91,"children":93},{"className":92},[],[94],{"type":55,"value":95},"I4H_WORKFLOWS",{"type":55,"value":97}," to its path (no clone happens). Otherwise this resolves the current repo, or clones to ",{"type":49,"tag":82,"props":99,"children":101},{"className":100},[],[102],{"type":55,"value":103},"~\u002Fi4h-workflows",{"type":55,"value":105}," — pick that default without prompting. Run every command below from the resolved root:",{"type":49,"tag":107,"props":108,"children":113},"pre",{"className":109,"code":110,"language":111,"meta":112,"style":112},"language-bash shiki shiki-themes material-theme-lighter material-theme material-theme-palenight","# Resolve the i4h-workflows base code (provides workflows\u002Fagentic\u002F).\nROOT=\"${I4H_WORKFLOWS:-$(git rev-parse --show-toplevel 2>\u002Fdev\u002Fnull)}\"\nif [ ! -d \"$ROOT\u002Fworkflows\u002Fagentic\" ]; then\n  ROOT=\"${I4H_WORKFLOWS:-$HOME\u002Fi4h-workflows}\"\n  [ -d \"$ROOT\u002Fworkflows\u002Fagentic\" ] || git clone https:\u002F\u002Fgithub.com\u002Fisaac-for-healthcare\u002Fi4h-workflows \"$ROOT\"\nfi\nexport I4H_WORKFLOWS=\"$ROOT\"; cd \"$ROOT\"\n","bash","",[114],{"type":49,"tag":82,"props":115,"children":116},{"__ignoreMap":112},[117,129,219,274,317,385,394],{"type":49,"tag":118,"props":119,"children":122},"span",{"class":120,"line":121},"line",1,[123],{"type":49,"tag":118,"props":124,"children":126},{"style":125},"--shiki-light:#90A4AE;--shiki-light-font-style:italic;--shiki-default:#546E7A;--shiki-default-font-style:italic;--shiki-dark:#676E95;--shiki-dark-font-style:italic",[127],{"type":55,"value":128},"# Resolve the i4h-workflows base code (provides workflows\u002Fagentic\u002F).\n",{"type":49,"tag":118,"props":130,"children":132},{"class":120,"line":131},2,[133,139,145,150,154,159,165,170,175,180,185,190,195,200,204,209,214],{"type":49,"tag":118,"props":134,"children":136},{"style":135},"--shiki-light:#90A4AE;--shiki-default:#EEFFFF;--shiki-dark:#BABED8",[137],{"type":55,"value":138},"ROOT",{"type":49,"tag":118,"props":140,"children":142},{"style":141},"--shiki-light:#39ADB5;--shiki-default:#89DDFF;--shiki-dark:#89DDFF",[143],{"type":55,"value":144},"=",{"type":49,"tag":118,"props":146,"children":147},{"style":141},[148],{"type":55,"value":149},"\"${",{"type":49,"tag":118,"props":151,"children":152},{"style":135},[153],{"type":55,"value":95},{"type":49,"tag":118,"props":155,"children":156},{"style":141},[157],{"type":55,"value":158},":-",{"type":49,"tag":118,"props":160,"children":162},{"style":161},"--shiki-light:#91B859;--shiki-default:#C3E88D;--shiki-dark:#C3E88D",[163],{"type":55,"value":164},"$(",{"type":49,"tag":118,"props":166,"children":167},{"style":135},[168],{"type":55,"value":169},"git",{"type":49,"tag":118,"props":171,"children":172},{"style":135},[173],{"type":55,"value":174}," rev-parse",{"type":49,"tag":118,"props":176,"children":177},{"style":135},[178],{"type":55,"value":179}," --show-toplevel",{"type":49,"tag":118,"props":181,"children":182},{"style":135},[183],{"type":55,"value":184}," 2",{"type":49,"tag":118,"props":186,"children":187},{"style":161},[188],{"type":55,"value":189},">",{"type":49,"tag":118,"props":191,"children":192},{"style":141},[193],{"type":55,"value":194},"\u002F",{"type":49,"tag":118,"props":196,"children":197},{"style":135},[198],{"type":55,"value":199},"dev",{"type":49,"tag":118,"props":201,"children":202},{"style":141},[203],{"type":55,"value":194},{"type":49,"tag":118,"props":205,"children":206},{"style":135},[207],{"type":55,"value":208},"null",{"type":49,"tag":118,"props":210,"children":211},{"style":161},[212],{"type":55,"value":213},")",{"type":49,"tag":118,"props":215,"children":216},{"style":141},[217],{"type":55,"value":218},"}\"\n",{"type":49,"tag":118,"props":220,"children":222},{"class":120,"line":221},3,[223,229,234,239,244,249,254,259,264,269],{"type":49,"tag":118,"props":224,"children":226},{"style":225},"--shiki-light:#39ADB5;--shiki-light-font-style:italic;--shiki-default:#89DDFF;--shiki-default-font-style:italic;--shiki-dark:#89DDFF;--shiki-dark-font-style:italic",[227],{"type":55,"value":228},"if",{"type":49,"tag":118,"props":230,"children":231},{"style":141},[232],{"type":55,"value":233}," [",{"type":49,"tag":118,"props":235,"children":236},{"style":141},[237],{"type":55,"value":238}," !",{"type":49,"tag":118,"props":240,"children":241},{"style":141},[242],{"type":55,"value":243}," -d",{"type":49,"tag":118,"props":245,"children":246},{"style":141},[247],{"type":55,"value":248}," \"",{"type":49,"tag":118,"props":250,"children":251},{"style":135},[252],{"type":55,"value":253},"$ROOT",{"type":49,"tag":118,"props":255,"children":256},{"style":161},[257],{"type":55,"value":258},"\u002Fworkflows\u002Fagentic",{"type":49,"tag":118,"props":260,"children":261},{"style":141},[262],{"type":55,"value":263},"\"",{"type":49,"tag":118,"props":265,"children":266},{"style":141},[267],{"type":55,"value":268}," ];",{"type":49,"tag":118,"props":270,"children":271},{"style":225},[272],{"type":55,"value":273}," then\n",{"type":49,"tag":118,"props":275,"children":277},{"class":120,"line":276},4,[278,283,287,291,295,299,304,308,313],{"type":49,"tag":118,"props":279,"children":280},{"style":135},[281],{"type":55,"value":282},"  ROOT",{"type":49,"tag":118,"props":284,"children":285},{"style":141},[286],{"type":55,"value":144},{"type":49,"tag":118,"props":288,"children":289},{"style":141},[290],{"type":55,"value":149},{"type":49,"tag":118,"props":292,"children":293},{"style":135},[294],{"type":55,"value":95},{"type":49,"tag":118,"props":296,"children":297},{"style":141},[298],{"type":55,"value":158},{"type":49,"tag":118,"props":300,"children":301},{"style":135},[302],{"type":55,"value":303},"$HOME",{"type":49,"tag":118,"props":305,"children":306},{"style":141},[307],{"type":55,"value":194},{"type":49,"tag":118,"props":309,"children":310},{"style":135},[311],{"type":55,"value":312},"i4h-workflows",{"type":49,"tag":118,"props":314,"children":315},{"style":141},[316],{"type":55,"value":218},{"type":49,"tag":118,"props":318,"children":320},{"class":120,"line":319},5,[321,326,330,334,338,342,346,351,356,362,367,372,376,380],{"type":49,"tag":118,"props":322,"children":323},{"style":141},[324],{"type":55,"value":325},"  [",{"type":49,"tag":118,"props":327,"children":328},{"style":141},[329],{"type":55,"value":243},{"type":49,"tag":118,"props":331,"children":332},{"style":141},[333],{"type":55,"value":248},{"type":49,"tag":118,"props":335,"children":336},{"style":135},[337],{"type":55,"value":253},{"type":49,"tag":118,"props":339,"children":340},{"style":161},[341],{"type":55,"value":258},{"type":49,"tag":118,"props":343,"children":344},{"style":141},[345],{"type":55,"value":263},{"type":49,"tag":118,"props":347,"children":348},{"style":141},[349],{"type":55,"value":350}," ]",{"type":49,"tag":118,"props":352,"children":353},{"style":141},[354],{"type":55,"value":355}," ||",{"type":49,"tag":118,"props":357,"children":359},{"style":358},"--shiki-light:#E2931D;--shiki-default:#FFCB6B;--shiki-dark:#FFCB6B",[360],{"type":55,"value":361}," git",{"type":49,"tag":118,"props":363,"children":364},{"style":161},[365],{"type":55,"value":366}," clone",{"type":49,"tag":118,"props":368,"children":369},{"style":161},[370],{"type":55,"value":371}," https:\u002F\u002Fgithub.com\u002Fisaac-for-healthcare\u002Fi4h-workflows",{"type":49,"tag":118,"props":373,"children":374},{"style":141},[375],{"type":55,"value":248},{"type":49,"tag":118,"props":377,"children":378},{"style":135},[379],{"type":55,"value":253},{"type":49,"tag":118,"props":381,"children":382},{"style":141},[383],{"type":55,"value":384},"\"\n",{"type":49,"tag":118,"props":386,"children":388},{"class":120,"line":387},6,[389],{"type":49,"tag":118,"props":390,"children":391},{"style":225},[392],{"type":55,"value":393},"fi\n",{"type":49,"tag":118,"props":395,"children":397},{"class":120,"line":396},7,[398,404,409,413,417,421,425,430,436,440,444],{"type":49,"tag":118,"props":399,"children":401},{"style":400},"--shiki-light:#9C3EDA;--shiki-default:#C792EA;--shiki-dark:#C792EA",[402],{"type":55,"value":403},"export",{"type":49,"tag":118,"props":405,"children":406},{"style":135},[407],{"type":55,"value":408}," I4H_WORKFLOWS",{"type":49,"tag":118,"props":410,"children":411},{"style":141},[412],{"type":55,"value":144},{"type":49,"tag":118,"props":414,"children":415},{"style":141},[416],{"type":55,"value":263},{"type":49,"tag":118,"props":418,"children":419},{"style":135},[420],{"type":55,"value":253},{"type":49,"tag":118,"props":422,"children":423},{"style":141},[424],{"type":55,"value":263},{"type":49,"tag":118,"props":426,"children":427},{"style":141},[428],{"type":55,"value":429},";",{"type":49,"tag":118,"props":431,"children":433},{"style":432},"--shiki-light:#6182B8;--shiki-default:#82AAFF;--shiki-dark:#82AAFF",[434],{"type":55,"value":435}," cd",{"type":49,"tag":118,"props":437,"children":438},{"style":141},[439],{"type":55,"value":248},{"type":49,"tag":118,"props":441,"children":442},{"style":135},[443],{"type":55,"value":253},{"type":49,"tag":118,"props":445,"children":446},{"style":141},[447],{"type":55,"value":384},{"type":49,"tag":58,"props":449,"children":451},{"id":450},"basics",[452],{"type":55,"value":453},"Basics",{"type":49,"tag":455,"props":456,"children":457},"ul",{},[458,502,507,512],{"type":49,"tag":459,"props":460,"children":461},"li",{},[462,468,470,476,478,484,486,492,494,500],{"type":49,"tag":463,"props":464,"children":465},"strong",{},[466],{"type":55,"value":467},"Env config (source of truth):",{"type":55,"value":469}," ",{"type":49,"tag":82,"props":471,"children":473},{"className":472},[],[474],{"type":55,"value":475},"workflows\u002Fagentic\u002Fconfig\u002Fenvironments\u002F\u003Cenv>.yaml",{"type":55,"value":477}," — drives every stage for ",{"type":49,"tag":82,"props":479,"children":481},{"className":480},[],[482],{"type":55,"value":483},"\u003Cenv>",{"type":55,"value":485}," (robot, task, policy, cameras, ",{"type":49,"tag":82,"props":487,"children":489},{"className":488},[],[490],{"type":55,"value":491},"arena.max_timesteps",{"type":55,"value":493},", ",{"type":49,"tag":82,"props":495,"children":497},{"className":496},[],[498],{"type":55,"value":499},"dataset.*",{"type":55,"value":501}," mappings).",{"type":49,"tag":459,"props":503,"children":504},{},[505],{"type":55,"value":506},"Use the e2e script for full pipeline runs.",{"type":49,"tag":459,"props":508,"children":509},{},[510],{"type":55,"value":511},"For per-stage work, use the corresponding dataset\u002Ffinetune\u002Fvalidate skills.",{"type":49,"tag":459,"props":513,"children":514},{},[515,521],{"type":49,"tag":82,"props":516,"children":518},{"className":517},[],[519],{"type":55,"value":520},"assemble_trocar",{"type":55,"value":522}," is inference-only; the e2e script skips finetune and checkpoint validation for it.",{"type":49,"tag":58,"props":524,"children":526},{"id":525},"dry-run",[527],{"type":55,"value":528},"Dry Run",{"type":49,"tag":107,"props":530,"children":532},{"className":109,"code":531,"language":111,"meta":112,"style":112},"REPO_ROOT=\"${I4H_WORKFLOWS:-$(git rev-parse --show-toplevel 2>\u002Fdev\u002Fnull)}\"; [ -d \"$REPO_ROOT\u002Fworkflows\u002Fagentic\" ] || REPO_ROOT=\"$HOME\u002Fi4h-workflows\"\n\"${REPO_ROOT}\u002Fworkflows\u002Fagentic\u002Fscripts\u002Fe2e\u002Frun.sh\" --dry-run --env \u003Cenv>\n",[533],{"type":49,"tag":82,"props":534,"children":535},{"__ignoreMap":112},[536,672],{"type":49,"tag":118,"props":537,"children":538},{"class":120,"line":121},[539,544,548,552,556,560,564,568,572,576,580,584,588,592,596,600,604,609,613,617,621,625,630,634,638,642,646,651,655,659,663,668],{"type":49,"tag":118,"props":540,"children":541},{"style":135},[542],{"type":55,"value":543},"REPO_ROOT",{"type":49,"tag":118,"props":545,"children":546},{"style":141},[547],{"type":55,"value":144},{"type":49,"tag":118,"props":549,"children":550},{"style":141},[551],{"type":55,"value":149},{"type":49,"tag":118,"props":553,"children":554},{"style":135},[555],{"type":55,"value":95},{"type":49,"tag":118,"props":557,"children":558},{"style":141},[559],{"type":55,"value":158},{"type":49,"tag":118,"props":561,"children":562},{"style":161},[563],{"type":55,"value":164},{"type":49,"tag":118,"props":565,"children":566},{"style":135},[567],{"type":55,"value":169},{"type":49,"tag":118,"props":569,"children":570},{"style":135},[571],{"type":55,"value":174},{"type":49,"tag":118,"props":573,"children":574},{"style":135},[575],{"type":55,"value":179},{"type":49,"tag":118,"props":577,"children":578},{"style":135},[579],{"type":55,"value":184},{"type":49,"tag":118,"props":581,"children":582},{"style":161},[583],{"type":55,"value":189},{"type":49,"tag":118,"props":585,"children":586},{"style":141},[587],{"type":55,"value":194},{"type":49,"tag":118,"props":589,"children":590},{"style":135},[591],{"type":55,"value":199},{"type":49,"tag":118,"props":593,"children":594},{"style":141},[595],{"type":55,"value":194},{"type":49,"tag":118,"props":597,"children":598},{"style":135},[599],{"type":55,"value":208},{"type":49,"tag":118,"props":601,"children":602},{"style":161},[603],{"type":55,"value":213},{"type":49,"tag":118,"props":605,"children":606},{"style":141},[607],{"type":55,"value":608},"}\"",{"type":49,"tag":118,"props":610,"children":611},{"style":141},[612],{"type":55,"value":429},{"type":49,"tag":118,"props":614,"children":615},{"style":141},[616],{"type":55,"value":233},{"type":49,"tag":118,"props":618,"children":619},{"style":141},[620],{"type":55,"value":243},{"type":49,"tag":118,"props":622,"children":623},{"style":141},[624],{"type":55,"value":248},{"type":49,"tag":118,"props":626,"children":627},{"style":135},[628],{"type":55,"value":629},"$REPO_ROOT",{"type":49,"tag":118,"props":631,"children":632},{"style":161},[633],{"type":55,"value":258},{"type":49,"tag":118,"props":635,"children":636},{"style":141},[637],{"type":55,"value":263},{"type":49,"tag":118,"props":639,"children":640},{"style":141},[641],{"type":55,"value":350},{"type":49,"tag":118,"props":643,"children":644},{"style":141},[645],{"type":55,"value":355},{"type":49,"tag":118,"props":647,"children":648},{"style":135},[649],{"type":55,"value":650}," REPO_ROOT",{"type":49,"tag":118,"props":652,"children":653},{"style":141},[654],{"type":55,"value":144},{"type":49,"tag":118,"props":656,"children":657},{"style":141},[658],{"type":55,"value":263},{"type":49,"tag":118,"props":660,"children":661},{"style":135},[662],{"type":55,"value":303},{"type":49,"tag":118,"props":664,"children":665},{"style":161},[666],{"type":55,"value":667},"\u002Fi4h-workflows",{"type":49,"tag":118,"props":669,"children":670},{"style":141},[671],{"type":55,"value":384},{"type":49,"tag":118,"props":673,"children":674},{"class":120,"line":131},[675,679,684,688,693,698,703,708,713,718,723],{"type":49,"tag":118,"props":676,"children":677},{"style":358},[678],{"type":55,"value":263},{"type":49,"tag":118,"props":680,"children":681},{"style":141},[682],{"type":55,"value":683},"${",{"type":49,"tag":118,"props":685,"children":686},{"style":135},[687],{"type":55,"value":543},{"type":49,"tag":118,"props":689,"children":690},{"style":141},[691],{"type":55,"value":692},"}",{"type":49,"tag":118,"props":694,"children":695},{"style":358},[696],{"type":55,"value":697},"\u002Fworkflows\u002Fagentic\u002Fscripts\u002Fe2e\u002Frun.sh\"",{"type":49,"tag":118,"props":699,"children":700},{"style":161},[701],{"type":55,"value":702}," --dry-run",{"type":49,"tag":118,"props":704,"children":705},{"style":161},[706],{"type":55,"value":707}," --env",{"type":49,"tag":118,"props":709,"children":710},{"style":141},[711],{"type":55,"value":712}," \u003C",{"type":49,"tag":118,"props":714,"children":715},{"style":161},[716],{"type":55,"value":717},"en",{"type":49,"tag":118,"props":719,"children":720},{"style":135},[721],{"type":55,"value":722},"v",{"type":49,"tag":118,"props":724,"children":725},{"style":141},[726],{"type":55,"value":727},">\n",{"type":49,"tag":58,"props":729,"children":731},{"id":730},"run",[732],{"type":55,"value":733},"Run",{"type":49,"tag":65,"props":735,"children":736},{},[737],{"type":55,"value":738},"Run the steps below in order. Each step is a separate bash call; variables persist in the local agent's tmux session.",{"type":49,"tag":65,"props":740,"children":741},{},[742,744,750,752,758,759,765,766,772,773,779,781,787,789,795],{"type":55,"value":743},"For Claude Code ",{"type":49,"tag":82,"props":745,"children":747},{"className":746},[],[748],{"type":55,"value":749},"--print",{"type":55,"value":751}," or any other noninteractive runner, keep Step 2 in the foreground. This is a validation requirement: do not use Claude background tasks, async task mode, Bash background mode, ",{"type":49,"tag":82,"props":753,"children":755},{"className":754},[],[756],{"type":55,"value":757},"&",{"type":55,"value":493},{"type":49,"tag":82,"props":760,"children":762},{"className":761},[],[763],{"type":55,"value":764},"nohup",{"type":55,"value":493},{"type":49,"tag":82,"props":767,"children":769},{"className":768},[],[770],{"type":55,"value":771},"tmux",{"type":55,"value":493},{"type":49,"tag":82,"props":774,"children":776},{"className":775},[],[777],{"type":55,"value":778},"disown",{"type":55,"value":780},", or any detached process\u002Ftask id, and do not answer that the pipeline is still running. Do not return until ",{"type":49,"tag":82,"props":782,"children":784},{"className":783},[],[785],{"type":55,"value":786},"run.sh",{"type":55,"value":788}," exits and you have inspected ",{"type":49,"tag":82,"props":790,"children":792},{"className":791},[],[793],{"type":55,"value":794},"logs\u002FSUMMARY.txt",{"type":55,"value":796}," on success, or the failing stage log on failure. Report the run dir, skipped stages, per-stage status, key artifacts, and cleanup\u002Fstop status before finishing.",{"type":49,"tag":798,"props":799,"children":801},"h3",{"id":800},"step-1-setup",[802],{"type":55,"value":803},"Step 1 — setup",{"type":49,"tag":107,"props":805,"children":807},{"className":109,"code":806,"language":111,"meta":112,"style":112},"REPO_ROOT=\"${I4H_WORKFLOWS:-$(git rev-parse --show-toplevel 2>\u002Fdev\u002Fnull)}\"; [ -d \"$REPO_ROOT\u002Fworkflows\u002Fagentic\" ] || REPO_ROOT=\"$HOME\u002Fi4h-workflows\"\n",[808],{"type":49,"tag":82,"props":809,"children":810},{"__ignoreMap":112},[811],{"type":49,"tag":118,"props":812,"children":813},{"class":120,"line":121},[814,818,822,826,830,834,838,842,846,850,854,858,862,866,870,874,878,882,886,890,894,898,902,906,910,914,918,922,926,930,934,938],{"type":49,"tag":118,"props":815,"children":816},{"style":135},[817],{"type":55,"value":543},{"type":49,"tag":118,"props":819,"children":820},{"style":141},[821],{"type":55,"value":144},{"type":49,"tag":118,"props":823,"children":824},{"style":141},[825],{"type":55,"value":149},{"type":49,"tag":118,"props":827,"children":828},{"style":135},[829],{"type":55,"value":95},{"type":49,"tag":118,"props":831,"children":832},{"style":141},[833],{"type":55,"value":158},{"type":49,"tag":118,"props":835,"children":836},{"style":161},[837],{"type":55,"value":164},{"type":49,"tag":118,"props":839,"children":840},{"style":135},[841],{"type":55,"value":169},{"type":49,"tag":118,"props":843,"children":844},{"style":135},[845],{"type":55,"value":174},{"type":49,"tag":118,"props":847,"children":848},{"style":135},[849],{"type":55,"value":179},{"type":49,"tag":118,"props":851,"children":852},{"style":135},[853],{"type":55,"value":184},{"type":49,"tag":118,"props":855,"children":856},{"style":161},[857],{"type":55,"value":189},{"type":49,"tag":118,"props":859,"children":860},{"style":141},[861],{"type":55,"value":194},{"type":49,"tag":118,"props":863,"children":864},{"style":135},[865],{"type":55,"value":199},{"type":49,"tag":118,"props":867,"children":868},{"style":141},[869],{"type":55,"value":194},{"type":49,"tag":118,"props":871,"children":872},{"style":135},[873],{"type":55,"value":208},{"type":49,"tag":118,"props":875,"children":876},{"style":161},[877],{"type":55,"value":213},{"type":49,"tag":118,"props":879,"children":880},{"style":141},[881],{"type":55,"value":608},{"type":49,"tag":118,"props":883,"children":884},{"style":141},[885],{"type":55,"value":429},{"type":49,"tag":118,"props":887,"children":888},{"style":141},[889],{"type":55,"value":233},{"type":49,"tag":118,"props":891,"children":892},{"style":141},[893],{"type":55,"value":243},{"type":49,"tag":118,"props":895,"children":896},{"style":141},[897],{"type":55,"value":248},{"type":49,"tag":118,"props":899,"children":900},{"style":135},[901],{"type":55,"value":629},{"type":49,"tag":118,"props":903,"children":904},{"style":161},[905],{"type":55,"value":258},{"type":49,"tag":118,"props":907,"children":908},{"style":141},[909],{"type":55,"value":263},{"type":49,"tag":118,"props":911,"children":912},{"style":141},[913],{"type":55,"value":350},{"type":49,"tag":118,"props":915,"children":916},{"style":141},[917],{"type":55,"value":355},{"type":49,"tag":118,"props":919,"children":920},{"style":135},[921],{"type":55,"value":650},{"type":49,"tag":118,"props":923,"children":924},{"style":141},[925],{"type":55,"value":144},{"type":49,"tag":118,"props":927,"children":928},{"style":141},[929],{"type":55,"value":263},{"type":49,"tag":118,"props":931,"children":932},{"style":135},[933],{"type":55,"value":303},{"type":49,"tag":118,"props":935,"children":936},{"style":161},[937],{"type":55,"value":667},{"type":49,"tag":118,"props":939,"children":940},{"style":141},[941],{"type":55,"value":384},{"type":49,"tag":798,"props":943,"children":945},{"id":944},"step-2-e2e-pipeline",[946],{"type":55,"value":947},"Step 2 — e2e pipeline",{"type":49,"tag":107,"props":949,"children":951},{"className":109,"code":950,"language":111,"meta":112,"style":112},"\"${REPO_ROOT}\u002Fworkflows\u002Fagentic\u002Fscripts\u002Fe2e\u002Frun.sh\" --env \u003Cenv>\n",[952],{"type":49,"tag":82,"props":953,"children":954},{"__ignoreMap":112},[955],{"type":49,"tag":118,"props":956,"children":957},{"class":120,"line":121},[958,962,966,970,974,978,982,986,990,994],{"type":49,"tag":118,"props":959,"children":960},{"style":358},[961],{"type":55,"value":263},{"type":49,"tag":118,"props":963,"children":964},{"style":141},[965],{"type":55,"value":683},{"type":49,"tag":118,"props":967,"children":968},{"style":135},[969],{"type":55,"value":543},{"type":49,"tag":118,"props":971,"children":972},{"style":141},[973],{"type":55,"value":692},{"type":49,"tag":118,"props":975,"children":976},{"style":358},[977],{"type":55,"value":697},{"type":49,"tag":118,"props":979,"children":980},{"style":161},[981],{"type":55,"value":707},{"type":49,"tag":118,"props":983,"children":984},{"style":141},[985],{"type":55,"value":712},{"type":49,"tag":118,"props":987,"children":988},{"style":161},[989],{"type":55,"value":717},{"type":49,"tag":118,"props":991,"children":992},{"style":135},[993],{"type":55,"value":722},{"type":49,"tag":118,"props":995,"children":996},{"style":141},[997],{"type":55,"value":727},{"type":49,"tag":58,"props":999,"children":1001},{"id":1000},"flags",[1002],{"type":55,"value":1003},"Flags",{"type":49,"tag":455,"props":1005,"children":1006},{},[1007,1037,1048],{"type":49,"tag":459,"props":1008,"children":1009},{},[1010,1016,1017,1023,1024,1030,1031],{"type":49,"tag":82,"props":1011,"children":1013},{"className":1012},[],[1014],{"type":55,"value":1015},"--skip-mimic",{"type":55,"value":493},{"type":49,"tag":82,"props":1018,"children":1020},{"className":1019},[],[1021],{"type":55,"value":1022},"--skip-annotate",{"type":55,"value":493},{"type":49,"tag":82,"props":1025,"children":1027},{"className":1026},[],[1028],{"type":55,"value":1029},"--skip-replay",{"type":55,"value":493},{"type":49,"tag":82,"props":1032,"children":1034},{"className":1033},[],[1035],{"type":55,"value":1036},"--skip-viz",{"type":49,"tag":459,"props":1038,"children":1039},{},[1040,1046],{"type":49,"tag":82,"props":1041,"children":1043},{"className":1042},[],[1044],{"type":55,"value":1045},"--from-stage \u003Cstage> --run-dir \u003Cexisting-run>",{"type":55,"value":1047}," resumes from a prior run.",{"type":49,"tag":459,"props":1049,"children":1050},{},[1051,1053,1059,1061,1066],{"type":55,"value":1052},"Policy record\u002Fverify stages open the sim window by default. Set ",{"type":49,"tag":82,"props":1054,"children":1056},{"className":1055},[],[1057],{"type":55,"value":1058},"ARENA_HEADLESS=1",{"type":55,"value":1060}," before ",{"type":49,"tag":82,"props":1062,"children":1064},{"className":1063},[],[1065],{"type":55,"value":786},{"type":55,"value":1067}," only when the user explicitly asks for headless\u002Fno-window execution.",{"type":49,"tag":65,"props":1069,"children":1070},{},[1071,1073,1079],{"type":55,"value":1072},"Stages: ",{"type":49,"tag":82,"props":1074,"children":1076},{"className":1075},[],[1077],{"type":55,"value":1078},"setup record mimic annotate replay convert viz finetune validate summary",{"type":55,"value":1080},".",{"type":49,"tag":58,"props":1082,"children":1084},{"id":1083},"outputs",[1085],{"type":55,"value":1086},"Outputs",{"type":49,"tag":65,"props":1088,"children":1089},{},[1090,1092,1098,1100,1106],{"type":55,"value":1091},"The script prints ",{"type":49,"tag":82,"props":1093,"children":1095},{"className":1094},[],[1096],{"type":55,"value":1097},"RUN_DIR",{"type":55,"value":1099}," and symlinks it to ",{"type":49,"tag":82,"props":1101,"children":1103},{"className":1102},[],[1104],{"type":55,"value":1105},"runs\u002F.latest",{"type":55,"value":1107},". Subdirs:",{"type":49,"tag":455,"props":1109,"children":1110},{},[1111,1137,1146,1155],{"type":49,"tag":459,"props":1112,"children":1113},{},[1114,1120,1122,1128,1130,1135],{"type":49,"tag":82,"props":1115,"children":1117},{"className":1116},[],[1118],{"type":55,"value":1119},"logs\u002F",{"type":55,"value":1121}," — per-stage logs, ",{"type":49,"tag":82,"props":1123,"children":1125},{"className":1124},[],[1126],{"type":55,"value":1127},"workflow.log",{"type":55,"value":1129}," (full teed output), and ",{"type":49,"tag":82,"props":1131,"children":1133},{"className":1132},[],[1134],{"type":55,"value":794},{"type":55,"value":1136}," (the final summary report)",{"type":49,"tag":459,"props":1138,"children":1139},{},[1140],{"type":49,"tag":82,"props":1141,"children":1143},{"className":1142},[],[1144],{"type":55,"value":1145},"data\u002F",{"type":49,"tag":459,"props":1147,"children":1148},{},[1149],{"type":49,"tag":82,"props":1150,"children":1152},{"className":1151},[],[1153],{"type":55,"value":1154},"lerobot\u002F",{"type":49,"tag":459,"props":1156,"children":1157},{},[1158,1164],{"type":49,"tag":82,"props":1159,"children":1161},{"className":1160},[],[1162],{"type":55,"value":1163},"checkpoint\u002F",{"type":55,"value":1165}," (trainable envs only)",{"type":49,"tag":58,"props":1167,"children":1169},{"id":1168},"monitor",[1170],{"type":55,"value":1171},"Monitor",{"type":49,"tag":65,"props":1173,"children":1174},{},[1175,1180,1182,1187],{"type":49,"tag":82,"props":1176,"children":1178},{"className":1177},[],[1179],{"type":55,"value":786},{"type":55,"value":1181}," runs every stage in the foreground and returns only when the whole pipeline ends, so track a long run from a ",{"type":49,"tag":463,"props":1183,"children":1184},{},[1185],{"type":55,"value":1186},"separate shell",{"type":55,"value":1188}," (do not expect to query it from the shell that launched it):",{"type":49,"tag":107,"props":1190,"children":1192},{"className":109,"code":1191,"language":111,"meta":112,"style":112},"tail -f \"${REPO_ROOT}\u002Fworkflows\u002Fagentic\u002Fruns\u002F.latest\u002Flogs\u002Fworkflow.log\"   # live per-stage progress\ncat    \"${REPO_ROOT}\u002Fworkflows\u002Fagentic\u002Fruns\u002F.latest\u002Flogs\u002FSUMMARY.txt\"     # final report (once DONE)\n",[1193],{"type":49,"tag":82,"props":1194,"children":1195},{"__ignoreMap":112},[1196,1236],{"type":49,"tag":118,"props":1197,"children":1198},{"class":120,"line":121},[1199,1204,1209,1214,1218,1222,1227,1231],{"type":49,"tag":118,"props":1200,"children":1201},{"style":358},[1202],{"type":55,"value":1203},"tail",{"type":49,"tag":118,"props":1205,"children":1206},{"style":161},[1207],{"type":55,"value":1208}," -f",{"type":49,"tag":118,"props":1210,"children":1211},{"style":141},[1212],{"type":55,"value":1213}," \"${",{"type":49,"tag":118,"props":1215,"children":1216},{"style":135},[1217],{"type":55,"value":543},{"type":49,"tag":118,"props":1219,"children":1220},{"style":141},[1221],{"type":55,"value":692},{"type":49,"tag":118,"props":1223,"children":1224},{"style":161},[1225],{"type":55,"value":1226},"\u002Fworkflows\u002Fagentic\u002Fruns\u002F.latest\u002Flogs\u002Fworkflow.log",{"type":49,"tag":118,"props":1228,"children":1229},{"style":141},[1230],{"type":55,"value":263},{"type":49,"tag":118,"props":1232,"children":1233},{"style":125},[1234],{"type":55,"value":1235},"   # live per-stage progress\n",{"type":49,"tag":118,"props":1237,"children":1238},{"class":120,"line":131},[1239,1244,1249,1253,1257,1262,1266],{"type":49,"tag":118,"props":1240,"children":1241},{"style":358},[1242],{"type":55,"value":1243},"cat",{"type":49,"tag":118,"props":1245,"children":1246},{"style":141},[1247],{"type":55,"value":1248},"    \"${",{"type":49,"tag":118,"props":1250,"children":1251},{"style":135},[1252],{"type":55,"value":543},{"type":49,"tag":118,"props":1254,"children":1255},{"style":141},[1256],{"type":55,"value":692},{"type":49,"tag":118,"props":1258,"children":1259},{"style":161},[1260],{"type":55,"value":1261},"\u002Fworkflows\u002Fagentic\u002Fruns\u002F.latest\u002Flogs\u002FSUMMARY.txt",{"type":49,"tag":118,"props":1263,"children":1264},{"style":141},[1265],{"type":55,"value":263},{"type":49,"tag":118,"props":1267,"children":1268},{"style":125},[1269],{"type":55,"value":1270},"     # final report (once DONE)\n",{"type":49,"tag":58,"props":1272,"children":1274},{"id":1273},"stop",[1275],{"type":55,"value":1276},"Stop",{"type":49,"tag":798,"props":1278,"children":1280},{"id":1279},"step-3-stop-if-needed",[1281],{"type":55,"value":1282},"Step 3 — stop (if needed)",{"type":49,"tag":107,"props":1284,"children":1286},{"className":109,"code":1285,"language":111,"meta":112,"style":112},"\"${REPO_ROOT}\u002Fworkflows\u002Fagentic\u002Fstop.sh\" all --env \u003Cenv>\n",[1287],{"type":49,"tag":82,"props":1288,"children":1289},{"__ignoreMap":112},[1290],{"type":49,"tag":118,"props":1291,"children":1292},{"class":120,"line":121},[1293,1297,1301,1305,1309,1314,1319,1323,1327,1331,1335],{"type":49,"tag":118,"props":1294,"children":1295},{"style":358},[1296],{"type":55,"value":263},{"type":49,"tag":118,"props":1298,"children":1299},{"style":141},[1300],{"type":55,"value":683},{"type":49,"tag":118,"props":1302,"children":1303},{"style":135},[1304],{"type":55,"value":543},{"type":49,"tag":118,"props":1306,"children":1307},{"style":141},[1308],{"type":55,"value":692},{"type":49,"tag":118,"props":1310,"children":1311},{"style":358},[1312],{"type":55,"value":1313},"\u002Fworkflows\u002Fagentic\u002Fstop.sh\"",{"type":49,"tag":118,"props":1315,"children":1316},{"style":161},[1317],{"type":55,"value":1318}," all",{"type":49,"tag":118,"props":1320,"children":1321},{"style":161},[1322],{"type":55,"value":707},{"type":49,"tag":118,"props":1324,"children":1325},{"style":141},[1326],{"type":55,"value":712},{"type":49,"tag":118,"props":1328,"children":1329},{"style":161},[1330],{"type":55,"value":717},{"type":49,"tag":118,"props":1332,"children":1333},{"style":135},[1334],{"type":55,"value":722},{"type":49,"tag":118,"props":1336,"children":1337},{"style":141},[1338],{"type":55,"value":727},{"type":49,"tag":58,"props":1340,"children":1342},{"id":1341},"prerequisites",[1343],{"type":55,"value":1344},"Prerequisites",{"type":49,"tag":455,"props":1346,"children":1347},{},[1348,1369,1382],{"type":49,"tag":459,"props":1349,"children":1350},{},[1351,1353,1359,1361,1367],{"type":55,"value":1352},"Workflow set up via [[i4h-workflow-setup]] (the ",{"type":49,"tag":82,"props":1354,"children":1356},{"className":1355},[],[1357],{"type":55,"value":1358},".venv",{"type":55,"value":1360}," must exist); ",{"type":49,"tag":82,"props":1362,"children":1364},{"className":1363},[],[1365],{"type":55,"value":1366},"setup",{"type":55,"value":1368}," is also the first pipeline stage.",{"type":49,"tag":459,"props":1370,"children":1371},{},[1372,1374,1380],{"type":55,"value":1373},"A valid ",{"type":49,"tag":82,"props":1375,"children":1377},{"className":1376},[],[1378],{"type":55,"value":1379},"--env",{"type":55,"value":1381}," name to drive the run.",{"type":49,"tag":459,"props":1383,"children":1384},{},[1385],{"type":55,"value":1386},"For per-stage work, use the corresponding dataset\u002Ffinetune\u002Fvalidate skills instead.",{"type":49,"tag":58,"props":1388,"children":1390},{"id":1389},"limitations",[1391],{"type":55,"value":1392},"Limitations",{"type":49,"tag":455,"props":1394,"children":1395},{},[1396,1405,1415],{"type":49,"tag":459,"props":1397,"children":1398},{},[1399,1404],{"type":49,"tag":82,"props":1400,"children":1402},{"className":1401},[],[1403],{"type":55,"value":520},{"type":55,"value":522},{"type":49,"tag":459,"props":1406,"children":1407},{},[1408,1413],{"type":49,"tag":82,"props":1409,"children":1411},{"className":1410},[],[1412],{"type":55,"value":1163},{"type":55,"value":1414}," outputs are produced for trainable envs only.",{"type":49,"tag":459,"props":1416,"children":1417},{},[1418,1420,1426,1428,1434],{"type":55,"value":1419},"Resuming requires both ",{"type":49,"tag":82,"props":1421,"children":1423},{"className":1422},[],[1424],{"type":55,"value":1425},"--from-stage \u003Cstage>",{"type":55,"value":1427}," and ",{"type":49,"tag":82,"props":1429,"children":1431},{"className":1430},[],[1432],{"type":55,"value":1433},"--run-dir \u003Cexisting-run>",{"type":55,"value":1080},{"type":49,"tag":58,"props":1436,"children":1438},{"id":1437},"troubleshooting",[1439],{"type":55,"value":1440},"Troubleshooting",{"type":49,"tag":455,"props":1442,"children":1443},{},[1444,1460,1483,1514],{"type":49,"tag":459,"props":1445,"children":1446},{},[1447,1452,1453,1458],{"type":49,"tag":463,"props":1448,"children":1449},{},[1450],{"type":55,"value":1451},"Error:",{"type":55,"value":469},{"type":49,"tag":82,"props":1454,"children":1456},{"className":1455},[],[1457],{"type":55,"value":1358},{"type":55,"value":1459}," not found \u002F module import fails - Cause: workflow not set up. Fix: run [[i4h-workflow-setup]] first.",{"type":49,"tag":459,"props":1461,"children":1462},{},[1463,1467,1469,1474,1476,1482],{"type":49,"tag":463,"props":1464,"children":1465},{},[1466],{"type":55,"value":1451},{"type":55,"value":1468}," env not recognized - Cause: wrong ",{"type":49,"tag":82,"props":1470,"children":1472},{"className":1471},[],[1473],{"type":55,"value":1379},{"type":55,"value":1475}," name. Fix: pass a valid env name; dry-run first with ",{"type":49,"tag":82,"props":1477,"children":1479},{"className":1478},[],[1480],{"type":55,"value":1481},"--dry-run --env \u003Cenv>",{"type":55,"value":1080},{"type":49,"tag":459,"props":1484,"children":1485},{},[1486,1490,1492,1498,1500,1506,1508,1513],{"type":49,"tag":463,"props":1487,"children":1488},{},[1489],{"type":55,"value":1451},{"type":55,"value":1491}," resume fails to find prior outputs - Cause: ",{"type":49,"tag":82,"props":1493,"children":1495},{"className":1494},[],[1496],{"type":55,"value":1497},"--from-stage",{"type":55,"value":1499}," used without a matching ",{"type":49,"tag":82,"props":1501,"children":1503},{"className":1502},[],[1504],{"type":55,"value":1505},"--run-dir",{"type":55,"value":1507},". Fix: pass ",{"type":49,"tag":82,"props":1509,"children":1511},{"className":1510},[],[1512],{"type":55,"value":1045},{"type":55,"value":1080},{"type":49,"tag":459,"props":1515,"children":1516},{},[1517,1521,1523,1529],{"type":49,"tag":463,"props":1518,"children":1519},{},[1520],{"type":55,"value":1451},{"type":55,"value":1522}," stale processes block a rerun - Cause: a previous pipeline session is still running. Fix: run ",{"type":49,"tag":82,"props":1524,"children":1526},{"className":1525},[],[1527],{"type":55,"value":1528},"stop.sh all --env \u003Cenv>",{"type":55,"value":1530}," before retrying.",{"type":49,"tag":58,"props":1532,"children":1534},{"id":1533},"final-response",[1535],{"type":55,"value":1536},"Final Response",{"type":49,"tag":65,"props":1538,"children":1539},{},[1540],{"type":55,"value":1541},"Report env, run dir, skipped stages, per-stage success\u002Ffailure, key artifact paths.",{"type":49,"tag":1543,"props":1544,"children":1545},"style",{},[1546],{"type":55,"value":1547},"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":1549,"total":1651},[1550,1567,1581,1595,1607,1622,1637],{"slug":1551,"name":1551,"fn":1552,"description":1553,"org":1554,"tags":1555,"stars":23,"repoUrl":24,"updatedAt":1566},"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},[1556,1559,1562,1563],{"name":1557,"slug":1558,"type":15},"Data Analysis","data-analysis",{"name":1560,"slug":1561,"type":15},"Data Engineering","data-engineering",{"name":9,"slug":8,"type":15},{"name":1564,"slug":1565,"type":15},"Performance","performance","2026-07-14T05:28:43.176466",{"slug":1568,"name":1568,"fn":1569,"description":1570,"org":1571,"tags":1572,"stars":23,"repoUrl":24,"updatedAt":1580},"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},[1573,1576,1579],{"name":1574,"slug":1575,"type":15},"Deployment","deployment",{"name":1577,"slug":1578,"type":15},"Infrastructure","infrastructure",{"name":9,"slug":8,"type":15},"2026-07-14T05:29:06.667109",{"slug":1582,"name":1582,"fn":1583,"description":1584,"org":1585,"tags":1586,"stars":23,"repoUrl":24,"updatedAt":1594},"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},[1587,1590,1591],{"name":1588,"slug":1589,"type":15},"Agents","agents",{"name":9,"slug":8,"type":15},{"name":1592,"slug":1593,"type":15},"Research","research","2026-07-14T05:28:06.816956",{"slug":1596,"name":1596,"fn":1597,"description":1598,"org":1599,"tags":1600,"stars":23,"repoUrl":24,"updatedAt":1606},"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},[1601,1602,1603],{"name":1557,"slug":1558,"type":15},{"name":9,"slug":8,"type":15},{"name":1604,"slug":1605,"type":15},"Testing","testing","2026-07-17T05:29:03.913266",{"slug":1608,"name":1608,"fn":1609,"description":1610,"org":1611,"tags":1612,"stars":23,"repoUrl":24,"updatedAt":1621},"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},[1613,1614,1617,1618],{"name":17,"slug":18,"type":15},{"name":1615,"slug":1616,"type":15},"Imaging","imaging",{"name":9,"slug":8,"type":15},{"name":1619,"slug":1620,"type":15},"Video","video","2026-07-17T05:28:53.905004",{"slug":1623,"name":1623,"fn":1624,"description":1625,"org":1626,"tags":1627,"stars":23,"repoUrl":24,"updatedAt":1636},"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},[1628,1629,1632,1633],{"name":1574,"slug":1575,"type":15},{"name":1630,"slug":1631,"type":15},"Docker","docker",{"name":9,"slug":8,"type":15},{"name":1634,"slug":1635,"type":15},"Operations","operations","2026-07-17T05:28:56.913999",{"slug":1638,"name":1638,"fn":1639,"description":1640,"org":1641,"tags":1642,"stars":23,"repoUrl":24,"updatedAt":1650},"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},[1643,1644,1647],{"name":9,"slug":8,"type":15},{"name":1645,"slug":1646,"type":15},"Quantum Computing","quantum-computing",{"name":1648,"slug":1649,"type":15},"Simulation","simulation","2026-07-14T05:26:58.898253",305,{"items":1653,"total":1802},[1654,1672,1688,1699,1711,1725,1738,1750,1761,1770,1784,1793],{"slug":1655,"name":1655,"fn":1656,"description":1657,"org":1658,"tags":1659,"stars":1669,"repoUrl":1670,"updatedAt":1671},"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},[1660,1663,1666],{"name":1661,"slug":1662,"type":15},"Documentation","documentation",{"name":1664,"slug":1665,"type":15},"MCP","mcp",{"name":1667,"slug":1668,"type":15},"Search","search",21777,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FNemoClaw","2026-07-20T06:00:01.461044",{"slug":1673,"name":1673,"fn":1674,"description":1675,"org":1676,"tags":1677,"stars":1685,"repoUrl":1686,"updatedAt":1687},"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},[1678,1681,1682],{"name":1679,"slug":1680,"type":15},"Containers","containers",{"name":1574,"slug":1575,"type":15},{"name":1683,"slug":1684,"type":15},"Python","python",17049,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FMegatron-LM","2026-07-27T06:06:11.249662",{"slug":1689,"name":1689,"fn":1690,"description":1691,"org":1692,"tags":1693,"stars":1685,"repoUrl":1686,"updatedAt":1698},"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},[1694,1697],{"name":1695,"slug":1696,"type":15},"CI\u002FCD","ci-cd",{"name":1574,"slug":1575,"type":15},"2026-07-14T05:25:59.97109",{"slug":1700,"name":1700,"fn":1701,"description":1702,"org":1703,"tags":1704,"stars":1685,"repoUrl":1686,"updatedAt":1710},"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},[1705,1706,1707],{"name":1695,"slug":1696,"type":15},{"name":1574,"slug":1575,"type":15},{"name":1708,"slug":1709,"type":15},"GitHub","github","2026-07-27T06:06:12.278222",{"slug":1712,"name":1712,"fn":1713,"description":1714,"org":1715,"tags":1716,"stars":1685,"repoUrl":1686,"updatedAt":1724},"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},[1717,1720,1721],{"name":1718,"slug":1719,"type":15},"Debugging","debugging",{"name":1708,"slug":1709,"type":15},{"name":1722,"slug":1723,"type":15},"Triage","triage","2026-07-14T05:25:57.442089",{"slug":1726,"name":1726,"fn":1727,"description":1728,"org":1729,"tags":1730,"stars":1685,"repoUrl":1686,"updatedAt":1737},"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},[1731,1734],{"name":1732,"slug":1733,"type":15},"Best Practices","best-practices",{"name":1735,"slug":1736,"type":15},"Code Analysis","code-analysis","2026-07-14T05:25:56.18433",{"slug":1739,"name":1739,"fn":1740,"description":1741,"org":1742,"tags":1743,"stars":1685,"repoUrl":1686,"updatedAt":1749},"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},[1744,1745,1748],{"name":20,"slug":21,"type":15},{"name":1746,"slug":1747,"type":15},"Migration","migration",{"name":9,"slug":8,"type":15},"2026-07-17T06:07:11.777011",{"slug":1751,"name":1751,"fn":1752,"description":1753,"org":1754,"tags":1755,"stars":1685,"repoUrl":1686,"updatedAt":1760},"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},[1756,1759],{"name":1757,"slug":1758,"type":15},"QA","qa",{"name":1604,"slug":1605,"type":15},"2026-07-14T05:25:53.673039",{"slug":1762,"name":1762,"fn":1763,"description":1764,"org":1765,"tags":1766,"stars":1685,"repoUrl":1686,"updatedAt":1769},"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},[1767,1768],{"name":1574,"slug":1575,"type":15},{"name":1577,"slug":1578,"type":15},"2026-07-14T05:25:49.362534",{"slug":1771,"name":1771,"fn":1772,"description":1773,"org":1774,"tags":1775,"stars":1685,"repoUrl":1686,"updatedAt":1783},"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},[1776,1779,1780],{"name":1777,"slug":1778,"type":15},"Code Review","code-review",{"name":1708,"slug":1709,"type":15},{"name":1781,"slug":1782,"type":15},"Pull Requests","pull-requests","2026-07-14T05:26:01.226578",{"slug":1785,"name":1785,"fn":1786,"description":1787,"org":1788,"tags":1789,"stars":1685,"repoUrl":1686,"updatedAt":1792},"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},[1790,1791],{"name":1757,"slug":1758,"type":15},{"name":1604,"slug":1605,"type":15},"2026-07-14T05:25:54.928983",{"slug":1794,"name":1794,"fn":1795,"description":1796,"org":1797,"tags":1798,"stars":1685,"repoUrl":1686,"updatedAt":1801},"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},[1799,1800],{"name":17,"slug":18,"type":15},{"name":1695,"slug":1696,"type":15},"2026-07-30T05:29:03.275638",496]