Sync from simready-oem-library-pm@5cf67ff7

README.md

@@ -1,44 +1,44 @@
----
-title: Simready Validator
-emoji: 🏢
-colorFrom: indigo
-colorTo: gray
-sdk: docker
-pinned: false
----
-
-# SimReady Validator
-
-A hosted validator for [SimReady](https://github.com/NVIDIA/simready-foundation)
-asset packages. Submit a HuggingFace dataset, pick a profile, and the Space
-runs the SimReady foundation rules against it and reports back what passed,
-what failed, and why — without copying your assets off HuggingFace.
-
-## Using the validator
-
-Click the **App** tab above to open the validator UI. In the form:
-
-| Field | What to enter |
-|---|---|
-| **Dataset** | The HuggingFace dataset slug you want to validate (e.g. `your-org/your-asset-bundle`). The dataset must be readable by this Space's `HF_TOKEN`. |
-| **Profile** | The SimReady profile that matches your asset type — props vs. robots, neutral vs. PhysX vs. Isaac. See the validation guide for the full list. |
-| **Version** | Profile version. Leave at `1.0.0` unless your profile is pinned to a different release. |
-| **Open PR on dataset with verdict** | When ticked, the verdict is written back to the dataset as a `validation/results.json` pull request. Leave off for ad-hoc smoke tests. |
-
-Click **Validate** and the log streams in real time. When it finishes the
-results.json appears in the **HTML report** download.
-
-## Going deeper
-
-- **[Partner walkthrough → VALIDATE.md](./blob/main/VALIDATE.md)** — the full nine-step guide from cloning the foundation repo to submitting your assets. Read this first if you're packaging assets for SimReady for the first time.
-- **[NVIDIA/simready-foundation](https://github.com/NVIDIA/simready-foundation)** — public spec repo with profile authoring guides and sample assets.
-- **[WRAPP on NGC](https://catalog.ngc.nvidia.com/orgs/nvidia/teams/omniverse/collections/wrapp)** — packaging tool referenced in the walkthrough.
-
-## About this Space
-
-The validator reads the requested HuggingFace dataset in place via
-`huggingface_hub.snapshot_download` — assets stay on HuggingFace, no
-transfer onto NVIDIA infrastructure. Running on a `t4-medium` GPU
-tier. PhysX / MDL rules that require an Isaac Sim runtime are not
-covered here yet (Isaac Sim isn't in the Space image — that's a
-separate workstream).
+---
+title: Simready Validator
+emoji: 🏢
+colorFrom: indigo
+colorTo: gray
+sdk: docker
+pinned: false
+---
+
+# SimReady Validator
+
+A hosted validator for [SimReady](https://github.com/NVIDIA/simready-foundation)
+asset packages. Submit a HuggingFace dataset, pick a profile, and the Space
+runs the SimReady foundation rules against it and reports back what passed,
+what failed, and why — without copying your assets off HuggingFace.
+
+## Using the validator
+
+Click the **App** tab above to open the validator UI. In the form:
+
+| Field | What to enter |
+|---|---|
+| **Dataset** | The HuggingFace dataset slug you want to validate (e.g. `your-org/your-asset-bundle`). The dataset must be readable by this Space's `HF_TOKEN`. |
+| **Profile** | The SimReady profile that matches your asset type — props vs. robots, neutral vs. PhysX vs. Isaac. See the validation guide for the full list. |
+| **Version** | Profile version. Leave at `1.0.0` unless your profile is pinned to a different release. |
+| **Open PR on dataset with verdict** | When ticked, the verdict is written back to the dataset as a `validation/results.json` pull request. Leave off for ad-hoc smoke tests. |
+
+Click **Validate** and the log streams in real time. When it finishes the
+results.json appears in the **HTML report** download.
+
+## Going deeper
+
+- **[Partner walkthrough → VALIDATE.md](./blob/main/VALIDATE.md)** — the full nine-step guide from cloning the foundation repo to submitting your assets. Read this first if you're packaging assets for SimReady for the first time.
+- **[NVIDIA/simready-foundation](https://github.com/NVIDIA/simready-foundation)** — public spec repo with profile authoring guides and sample assets.
+- **[WRAPP on NGC](https://catalog.ngc.nvidia.com/orgs/nvidia/teams/omniverse/collections/wrapp)** — packaging tool referenced in the walkthrough.
+
+## About this Space
+
+The validator reads the requested HuggingFace dataset in place via
+`huggingface_hub.snapshot_download` — assets stay on HuggingFace, no
+transfer onto NVIDIA infrastructure. Running on a `t4-medium` GPU
+tier. PhysX / MDL rules that require an Isaac Sim runtime are not
+covered here yet (Isaac Sim isn't in the Space image — that's a
+separate workstream).

VALIDATE.md

@@ -1,130 +1,130 @@
-# SimReady OEM Library — Onboarding
-
-## Step 1: Clone the Foundation Repo
-
-```bash
-git lfs install
-git clone https://github.com/NVIDIA/simready-foundation.git
-cd simready-foundation
-```
-
-> Requires [Git LFS](https://git-lfs.com). Run `git lfs pull` if binary files show as pointers.
-
----
-
-## Step 2: Understand Profiles
-
-Profiles define what your asset must satisfy. Pick the one matching your use case:
-
-| Profile | Use Case |
-|---------|----------|
-| `Prop-Robotics-Neutral` | Props (format-neutral physics) |
-| `Prop-Robotics-Physx` | Props (PhysX-specific) |
-| `Prop-Robotics-Isaac` | Props (Isaac Sim ready) |
-| `Robot-Body-Neutral` | Robot bodies (neutral) |
-| `Robot-Body-Runnable` | Robots runnable in PhysX |
-| `Robot-Body-Isaac` | Robots for Isaac Sim |
-
-> ⚠️ Profile names are **case-sensitive**.
-
-**Read your profile's authoring guide:**
-`nv_core/sr_specs/docs/profiles/<profile-name>.md`
-
-**See formal definitions:**
-`nv_core/sr_specs/docs/profiles/profiles.toml`
-
-**Study sample assets:**
-`sample_content/common_assets/robots_general/` and `props_general/`
-
----
-
-## Step 3: Install the SDK
-
-Install the `simready` CLI from the public SimReady OEM SDK distribution.
-
-```bash
-pip install simready-oem-sdk
-```
-
-> Avoid Anaconda — it causes USD DLL conflicts. The packaging spec is summarized in Step 4 below.
-
----
-
-## Step 4: Package Your Assets
-
-Follow the SimReady packaging specification (summarized below). Required structure:
-
-```
-<asset_name>/
-├── <asset_name>.usd           ← Interface file (validate THIS)
-├── configuration/
-│   ├── <asset_name>_base.usd
-│   ├── <asset_name>_physics.usd
-│   └── <asset_name>_sensor.usd
-└── .thumbs/256x256/
-    └── <asset_name>.usd.png   ← 256×256 thumbnail (REQUIRED)
-```
-
-Embed profile metadata in your interface file's `customLayerData`. See your profile's authoring guide under `nv_core/sr_specs/docs/profiles/<profile-name>.md` in the [NVIDIA/simready-foundation](https://github.com/NVIDIA/simready-foundation) repo for full details.
-
----
-
-## Step 5: Validate & Fix
-
-```bash
-simready validate ./my_asset/my_asset.usd --profile Robot-Body-Runnable
-simready fix list                       # see available auto-fixes
-simready fix apply <fix_name> ./my_asset
-```
-
-Iterate until all checks pass. For a hosted run that doesn't require local installation, use the **Validator** tab on this Space.
-
----
-
-## Step 6: Generate a Validation Receipt
-
-```bash
-simready validate ./my_asset/my_asset.usd --profile Robot-Body-Runnable --receipt
-```
-
-Creates `.simready/validation.json` — **required for submission**.
-
----
-
-## Step 7: Package with WRAPP (Optional)
-
-```bash
-wrapp create my_asset 1.0.0 ./my_asset --repo s3://your-bucket/assets
-```
-
-> WRAPP is from [NGC](https://catalog.ngc.nvidia.com/orgs/nvidia/teams/omniverse/collections/wrapp), not pip.
-
----
-
-## Step 8: Create Manifest & Pre-flight
-
-```bash
-simready manifest init my_company --email team@company.com
-simready manifest add my_company ./my_asset
-simready manifest validate my_company    # must pass before submission
-```
-
----
-
-## Step 9: Submit Merge Request
-
-```bash
-simready submit my_company
-git checkout -b submission/my_company-q2-2026
-git add . && git commit -m "Submit assets" && git push origin HEAD
-```
-
-Open a PR targeting `main`. Include your manifest + asset directories with receipts.
-
----
-
-## Further Reading
-
-- [NVIDIA/simready-foundation](https://github.com/NVIDIA/simready-foundation) — public specs, profile authoring guides, sample assets
-- [WRAPP on NGC](https://catalog.ngc.nvidia.com/orgs/nvidia/teams/omniverse/collections/wrapp) — packaging tool
-- [Git LFS](https://git-lfs.com) — required for the foundation repo's binary assets
+# SimReady OEM Library — Onboarding
+
+## Step 1: Clone the Foundation Repo
+
+```bash
+git lfs install
+git clone https://github.com/NVIDIA/simready-foundation.git
+cd simready-foundation
+```
+
+> Requires [Git LFS](https://git-lfs.com). Run `git lfs pull` if binary files show as pointers.
+
+---
+
+## Step 2: Understand Profiles
+
+Profiles define what your asset must satisfy. Pick the one matching your use case:
+
+| Profile | Use Case |
+|---------|----------|
+| `Prop-Robotics-Neutral` | Props (format-neutral physics) |
+| `Prop-Robotics-Physx` | Props (PhysX-specific) |
+| `Prop-Robotics-Isaac` | Props (Isaac Sim ready) |
+| `Robot-Body-Neutral` | Robot bodies (neutral) |
+| `Robot-Body-Runnable` | Robots runnable in PhysX |
+| `Robot-Body-Isaac` | Robots for Isaac Sim |
+
+> ⚠️ Profile names are **case-sensitive**.
+
+**Read your profile's authoring guide:**
+`nv_core/sr_specs/docs/profiles/<profile-name>.md`
+
+**See formal definitions:**
+`nv_core/sr_specs/docs/profiles/profiles.toml`
+
+**Study sample assets:**
+`sample_content/common_assets/robots_general/` and `props_general/`
+
+---
+
+## Step 3: Install the SDK
+
+Install the `simready` CLI from the public SimReady OEM SDK distribution.
+
+```bash
+pip install simready-oem-sdk
+```
+
+> Avoid Anaconda — it causes USD DLL conflicts. The packaging spec is summarized in Step 4 below.
+
+---
+
+## Step 4: Package Your Assets
+
+Follow the SimReady packaging specification (summarized below). Required structure:
+
+```
+<asset_name>/
+├── <asset_name>.usd           ← Interface file (validate THIS)
+├── configuration/
+│   ├── <asset_name>_base.usd
+│   ├── <asset_name>_physics.usd
+│   └── <asset_name>_sensor.usd
+└── .thumbs/256x256/
+    └── <asset_name>.usd.png   ← 256×256 thumbnail (REQUIRED)
+```
+
+Embed profile metadata in your interface file's `customLayerData`. See your profile's authoring guide under `nv_core/sr_specs/docs/profiles/<profile-name>.md` in the [NVIDIA/simready-foundation](https://github.com/NVIDIA/simready-foundation) repo for full details.
+
+---
+
+## Step 5: Validate & Fix
+
+```bash
+simready validate ./my_asset/my_asset.usd --profile Robot-Body-Runnable
+simready fix list                       # see available auto-fixes
+simready fix apply <fix_name> ./my_asset
+```
+
+Iterate until all checks pass. For a hosted run that doesn't require local installation, use the **Validator** tab on this Space.
+
+---
+
+## Step 6: Generate a Validation Receipt
+
+```bash
+simready validate ./my_asset/my_asset.usd --profile Robot-Body-Runnable --receipt
+```
+
+Creates `.simready/validation.json` — **required for submission**.
+
+---
+
+## Step 7: Package with WRAPP (Optional)
+
+```bash
+wrapp create my_asset 1.0.0 ./my_asset --repo s3://your-bucket/assets
+```
+
+> WRAPP is from [NGC](https://catalog.ngc.nvidia.com/orgs/nvidia/teams/omniverse/collections/wrapp), not pip.
+
+---
+
+## Step 8: Create Manifest & Pre-flight
+
+```bash
+simready manifest init my_company --email team@company.com
+simready manifest add my_company ./my_asset
+simready manifest validate my_company    # must pass before submission
+```
+
+---
+
+## Step 9: Submit Merge Request
+
+```bash
+simready submit my_company
+git checkout -b submission/my_company-q2-2026
+git add . && git commit -m "Submit assets" && git push origin HEAD
+```
+
+Open a PR targeting `main`. Include your manifest + asset directories with receipts.
+
+---
+
+## Further Reading
+
+- [NVIDIA/simready-foundation](https://github.com/NVIDIA/simready-foundation) — public specs, profile authoring guides, sample assets
+- [WRAPP on NGC](https://catalog.ngc.nvidia.com/orgs/nvidia/teams/omniverse/collections/wrapp) — packaging tool
+- [Git LFS](https://git-lfs.com) — required for the foundation repo's binary assets

tools/hf_space/README_HEADER.md

@@ -38,6 +38,7 @@ results.json appears in the **HTML report** download.
 
 The validator reads the requested HuggingFace dataset in place via
 `huggingface_hub.snapshot_download` — assets stay on HuggingFace, no
-transfer onto NVIDIA infrastructure. Hardware is `cpu-upgrade`
-(8 vCPU, 32 GB); Kit-rooted PhysX / MDL rule coverage requires a GPU
-tier and isn't enabled here yet.
+transfer onto NVIDIA infrastructure. Running on a `t4-medium` GPU
+tier. PhysX / MDL rules that require an Isaac Sim runtime are not
+covered here yet (Isaac Sim isn't in the Space image — that's a
+separate workstream).