Agent Native

simready-foundation-conform-fet-000-newton

Repairs Newton physics runtime variant scaffolding on USD assets by adding the variant set, payload arc, and metadata required for FET_000_NEWTON conformance.

by NVIDIAOfficial publisher

Identity proven by GitHub's verified organisation badge (NVIDIA, NVIDIA-NeMo)

Agents
Claude Code, Codex, Cursor, Gemini CLI, GitHub Copilot, OpenCode, claude.ai and API
License
Apache-2.0 (permissive)
Pin
60b8bddpinned Oct 9, 2026
Context cost
about 4,700 tokenswhen the skill loads
GitHub stars
95on nvidia/simready-foundation
Scripts
None found

What it does

“Use for repairing exact FET_000_NEWTON SimReady conformance: the Newton physics runtime variant scaffolding (Newton variant set, runnables/physics/newton payload, and SimReady variant metadata). Use when a profile, validation report, or user request names FET_000_NEWTON; default to version `0.1.0`.”
The author's description, from SKILL.md

Generated summary from the skill's own text

Repairs Newton physics runtime variant scaffolding on USD assets by adding the variant set, payload arc, and metadata required for FET_000_NEWTON conformance. It handles the structural container for Newton physics without authoring the physics itself, applying deterministic fixes for variant sets, payload paths, and isolation rules.

When to use it

  • Asset validation report names FET_000_NEWTON or variant structure is missing
  • Adding Newton runtime option to multi-runtime profiles like Robotics-Prop
  • Staging Newton variant scaffolding on single-mesh or multibody props
  • Ensuring payload anchors stay within asset root and isolation is pure

When not to

  • Asset still fails neutral Core conformance; repair FET_000_STANDARD first
  • Physics data, joints, or articulation authoring needed; use FET_003_NEWTON instead
  • PhysX or MuJoCo variants are the target; use their dedicated skills

Install

These commands install from the publisher's repository and need no account.

skills CLI

npx skills add nvidia/simready-foundation --skill simready-foundation-conform-fet-000-newton

Installs from the repository's default branch, which may have changed since the commit we checked.

GitHub CLI 2.90 or later

gh skill install nvidia/simready-foundation simready-foundation-conform-fet-000-newton@60b8bddb086f43c8cc394f50635b4e65b0ec36fd

Installs the commit we checked (60b8bdd).

Copy by hand

Download the folder at 60b8bdd and put it in .claude/skills/simready-foundation-conform-fet-000-newton for Claude Code and in .agents/skills/simready-foundation-conform-fet-000-newton for Codex, Cursor, Gemini CLI, GitHub Copilot and OpenCode.

Skill folder on GitHub at 60b8bdd

Claude Code plugin

/plugin install nvidia-skills@claude-plugins-official

The publisher's plugin in Claude's official marketplace. It may hold more than this skill, and we have not pinned it.

Copy for your agent

A prompt for any coding agent. It installs this skill from GitHub at the pinned commit, copies the files as they are, and runs nothing unless you ask.

Prompt

Install the agent skill "simready-foundation-conform-fet-000-newton" by NVIDIA.

Source, pinned to commit 60b8bddb086f43c8cc394f50635b4e65b0ec36fd:
https://github.com/nvidia/simready-foundation/tree/60b8bddb086f43c8cc394f50635b4e65b0ec36fd/skills/simready-foundation-conform-fet-000-newton

1. Download that folder at exactly this commit. Do not use the default branch.
2. Copy every file unchanged, keeping the folder structure, into .claude/skills/simready-foundation-conform-fet-000-newton/ for Claude Code and .agents/skills/simready-foundation-conform-fet-000-newton/ for Codex, Cursor, Gemini CLI, GitHub Copilot and OpenCode. Create the folders if they do not exist.
3. Do not edit the files, and do not run any script in the folder unless I ask you to.
4. When you are done, list the files you copied and where you put them.

Listing and checks: https://www.agentnative.dev/skills/nvidia/simready-foundation-conform-fet-000-newton

SKILL.md at the pin

---
name: simready-foundation-conform-fet-000-newton
description: "Use for repairing exact FET_000_NEWTON SimReady conformance: the Newton physics runtime variant scaffolding (Newton variant set, runnables/physics/newton payload, and SimReady variant metadata). Use when a profile, validation report, or user request names FET_000_NEWTON; default to version `0.1.0`."
license: Apache-2.0
metadata:
  author: "Shaad Boochoon <sboochoon@nvidia.com>"
  tags:
    - simready
    - conformance
    - newton
    - core
---

# SimReady Conform FET_000_NEWTON

## Purpose

Use this exact feature skill when the selected profile, validation report, or user request names `FET_000_NEWTON`. It repairs or stages the Newton physics runtime variant contract on top of neutral Core, without drifting into another runtime contract.

`FET_000_NEWTON` adds the Newton runtime variant scaffolding to an asset that already satisfies neutral Core (`FET_000_STANDARD@0.1.0`). It does not author Newton rigid-body/collider physics itself; that belongs to `FET_003_NEWTON` and related features. This skill only makes the `Newton` variant set, its `runnables/physics/newton` payload, and its SimReady variant metadata correct.

Default to `FET_000_NEWTON@0.1.0` when the user asks for this feature without a version. If the report names a different `FET_000_<RUNTIME>` feature (`FET_000_PHYSX`, `FET_000_MUJOCO`), switch to that feature's matching skill before editing.

To conform an asset to a profile like `Robotics-Prop@3.1.0` that lists all three optional runtime variants, run all three `FET_000_*` runtime skills so each of `PhysX`, `Newton`, and `MuJoCo` ends up with **both** a variant set on the default prim **and** a `SimReady_Metadata.Variants.Physics` entry. Never leave a `runnables/physics/<stem>.usd` payload on disk without its matching variant set and metadata — an orphaned payload fails `RV.007`/`RV.008`/`RV.009` for its runtime.

This skill is not the final validator and should not silently mutate source assets. It stages a repaired USD-family asset under the requested output directory, applies deterministic fixes only where safe, and stops when a validation gate still fails or a manual decision is required.

## Source of Truth

Before changing an asset or package, read:

- `nv_core/tiers/simready_foundation_tier_core/simready/foundation/tier_core/features/FET_000_NEWTON-0.1.0.json`
- `nv_core/tiers/simready_foundation_tier_core/simready/foundation/tier_core/features/FET_000_NEWTON.md`
- `nv_core/tiers/simready_foundation_tier_core/simready/foundation/tier_core/features/FET_000_STANDARD-0.1.0.json` (dependency)
- Requirement docs for the failing IDs under `nv_core/tiers/simready_foundation_tier_core/simready/foundation/tier_core/capabilities/core/runtime_variants/requirements/`:
  - `newton-variant-set.md` (RV.004)
  - `newton-runtime-payload.md` (RV.005)
  - `newton-variant-metadata.md` (RV.006)
  - `runtime-variant-section-purity.md` (RV.010)
  - `runtime-physics-isolation.md` (RV.011)
- For the cross-runtime isolation policy that RV.010/RV.011 enforce (including the strict Newton isolation rule), read
  `nv_core/tiers/simready_foundation_tier_core/simready/foundation/tier_core/capabilities/core/runtime_variants/runtime-physics-isolation-matrix.md`.
- For the Newton payload body (mesh collision authoring), read
  `nv_core/tiers/simready_foundation_tier_core/simready/foundation/tier_core/capabilities/physics_bodies/physics_rigid_bodies/requirements/newton-collider-api.md`
  and `nv_core/tiers/simready_foundation_tier_core/simready/foundation/tier_core/features/FET_003_NEWTON.md`.

Treat the selected JSON manifest as authoritative for dependencies and requirement IDs.

This skill is **self-contained**: the deterministic guidance and worked example below are sufficient to author a conforming `Newton` runtime variant from scratch. You do not need to open an existing 3.1.0 asset and copy its structure — do so only to cross-check, and only against a sample known to pass.

Sample-asset caveats (do not copy structure blindly):

- Author from the requirement docs and the worked example below, not by copying an existing asset. Sample assets drift, and a stale or pre-contract sample will lead you astray. If you do cross-check, only trust a sample you have **just** confirmed passes the full target profile (RV.001–011 **and** AA.001, not the RV-only harness) — never treat any specific asset as a canonical template.

## Feature Versions

| Version | Dependencies | Requirements |
|---|---|---|
| `0.1.0` | FET_000_STANDARD@0.1.0 | `RV.004`, `RV.005`, `RV.006`, `RV.010`, `RV.011` |

## Inputs

| Input | Requirement |
|---|---|
| `usd_asset` | Required USD-family asset to repair. |
| `output_root` | Required or inferred folder for staged assets and reports. |
| `simready_profile` / `profile_version` | Validation target, if supplied. |
| `validation_report` | Preferred JSON/markdown report from the failing gate. |
| `newton_overrides` | Optional source of the Newton runtime override data (schemas/attributes) to place in the payload; required to fully author `newton.usd` content. |

## Workflow

1. Confirm the input exists and identify the exact selected feature/version.
2. Ensure the neutral Core gate (`FET_000_STANDARD@0.1.0`) is in shape first; if Core fails, hand off to `simready-foundation-conform-fet-000-standard` before this skill.
3. Load the `FET_000_NEWTON` manifest, the feature markdown, and the RV.004-RV.006 + RV.010/RV.011 requirement docs (plus the isolation matrix).
4. Create or use a staged output location unless the user explicitly asks for in-place edits.
5. Repair only `RV.004`, `RV.005`, `RV.006`, `RV.010`, and `RV.011` (see Feature Guidance). Do not touch PhysX/MuJoCo variants or rigid-body physics.
6. Rerun the same profile gate or the narrowest available feature/capability validation gate.
7. Summarize the selected version, changed files, validation evidence, and the first remaining blocker or next exact feature gate.

## Feature Guidance

Repair each Newton runtime variant requirement deterministically where safe:

- `RV.004` (variant set): On the stage `defaultPrim`, add a `Newton` variant set with `Disabled` and `Enabled` options and author the default selection as `Disabled`. Keep the `Disabled` variant free of runtime payloads.
- `RV.005` (payload): From the `Enabled` variant, `prepend payload = @./runnables/physics/newton.usd@` (or `.usda`). The payload must live **inside the asset root** — the directory that holds the root layer — so keep `runnables/physics/` as a subfolder beside the root layer and anchor the arc with `./`. Do **not** place `runnables/` beside or above the asset root and reference it with `@../runnables/physics/newton.usd@`: a `../` payload that resolves above the asset root fails **AA.001** (atomic-asset anchored paths) and precludes `FET_001_STANDARD`, even though the RV.005 payload check alone still passes — so the RV-only local harness will not catch it. Create the `runnables/physics/newton.usd` layer as an `over` layer with the same `defaultPrim`. `.usd` and `.usda` are interchangeable; author `.usda` when a binary USDC cannot be compiled.
- `RV.006` (metadata): Add `customLayerData.SimReady_Metadata.Variants.Physics.Newton = {prim = <defaultPrim path>, variantSetName = "Newton", activateOption = "Enabled"}`. The `Variants` dict must live **inside** `SimReady_Metadata`, not as a sibling top-level `customLayerData.Variants` key.
- `RV.010` (variant-section purity): Keep the variant sections pure. The `Enabled` option must contain **only** the `prepend payload` arc — no inline `over`/`def` child prims, no authored properties, no other composition arcs. The `Disabled` option must be **completely empty**. Inline variant opinions outrank payloads in USD strength ordering (LIVRPS), so any override authored in a variant section leaks into the composed result of the *other* runtimes. All runtime data belongs in the `runnables/physics/newton.usd` payload, never in the root variant edit.
- `RV.011` (composed-stage isolation): The composed Newton stage (Newton `Enabled`, others `Disabled`) must carry only neutral `UsdPhysics`/`physics:*` data plus Newton data (`Newton*` schemas, `newton:*` attributes). **Strict Newton isolation:** do not author MuJoCo (`Mjc*`/`mjc:*`) or PhysX (`Physx*`/`physx*:`) schemas/attributes in `newton.usd`, even though `MjcCollisionAPI` inherits `NewtonCollisionAPI` upstream — the inheritance is one-directional and does not license a Newton layer to carry the more-specific MuJoCo schema. Do **not** author `physics:approximation = "sdf"` under Newton (that is a PhysX approximation and is a conflict/failure); Newton does not consume `physics:approximation`, so an authored value is at best an inert leak — the `newton.usd` payload should author **no** `physics:approximation` at all. RV.010/RV.011 are composition-wide checks: if a sibling runtime's section is impure or its payload leaks a foreign schema, report it rather than silently editing another runtime's contract.
- **Neutral-base precondition (RV.011).** Newton selects its collision representation from `NewtonMeshCollisionAPI`/`NewtonSDFCollisionAPI`, not from `physics:approximation`, so the neutral base collider must **not** author `physics:approximation` as a local (non-variant) opinion. A local base opinion outranks the payload arc in USD strength ordering (LIVRPS: Local > … > Payload), so a base `convexDecomposition`/`convexHull` composes into the Newton selection as an inert-leak warning, and a base `sdf` fails. If the neutral base authors `physics:approximation`, clear that local opinion so each runnable owns its own value (PhysX `sdf`, MuJoCo `convexHull`, Newton none). See the "Value-level attribute rules" in the isolation matrix.
- **Migrating a runtime-authored base (RV.011).** If the asset arrives with a different runtime authored directly on the neutral base — common when a prop was previously conformed to a single-runtime PhysX profile — the base collider carries that runtime's API schemas (e.g. `PhysxCollisionAPI`, `PhysxSDFMeshCollisionAPI`) alongside `PhysicsMeshCollisionAPI` and a local `physics:approximation`. Move that runtime data into its own runnable payload and reduce the neutral base collider to neutral `UsdPhysics` only (`PhysicsCollisionAPI` + `PhysicsMassAPI`), with the mesh-collision schema and approximation living in the per-runtime payloads. Otherwise the neutral (all-`Disabled`) composition still carries foreign runtime schemas and fails RV.011.

### Newton payload body for a single-mesh prop

For a simple prop whose collision is a single existing mesh, the `newton.usd` payload body is a deterministic mesh-collision delta. Apply on the collision mesh `over` (strict Newton — no `Mjc*`/`mjc:*`, and no `physics:approximation`):

- `prepend apiSchemas = ["NewtonCollisionAPI", "PhysicsMeshCollisionAPI", "NewtonMeshCollisionAPI"]`
- `float newton:contactGap = 0`

Do not create geometry; only add schemas/attributes to the mesh(es) that already exist in the neutral base.

Block and report instead of guessing when:

- The asset needs Newton joints, articulation, or scene tuning, or per-link collider decisions that are not the uniform delta below (e.g. mixing mesh and SDF colliders per link). Author the variant/payload scaffolding and hand that body work off to `FET_003_NEWTON`/`FET_004_NEWTON`/`FET_022_NEWTON` or the supplied `newton_overrides`. A plain multibody prop whose colliders all take the same delta is in scope — see "Multibody / multi-collider assemblies" below.
- The collision geometry or intended collider mesh is ambiguous.
- The asset has no `defaultPrim`, or the intended owning prim is ambiguous.

Keep joint/articulation/scene Newton physics work in `FET_003_NEWTON`/`FET_004_NEWTON`; this skill owns the runtime variant scaffolding plus the collision (and Newton material) deltas — one collider mesh for a single-mesh prop, or every collider mesh and bound material for a multibody assembly (below).

### Multibody / multi-collider assemblies

When the neutral base is a multibody assembly — several rigid-body parts sharing one articulation (e.g. a toolbox with a hinged lid, handle, and locks) — the `newton.usd` payload is the same collision delta applied to **every** collider, plus material tuning on every bound material:

- Apply the mesh-collision `over` above (`NewtonCollisionAPI`, `PhysicsMeshCollisionAPI`, `NewtonMeshCollisionAPI` + `float newton:contactGap = 0`) to **each** collider mesh in the assembly.
- On **each** bound physics material (a `UsdShade.Material` with `PhysicsMaterialAPI`), add an `over` with `prepend apiSchemas = ["NewtonMaterialAPI"]` and the default frictions `float newton:rollingFriction = 0.0001` / `float newton:torsionalFriction = 0.005` (placement only; `newton:*` values are trusted — NEWTON.MAT.001 aligns with PMT.001).
- Keep the payload collider/material-only: never re-author joints, articulation, or rigid bodies in `newton.usd` — that topology lives once on the neutral base.

Neutral-base preconditions for a multibody assembly (owned by `FET_004_STANDARD`/`FET_004_NEWTON`, repair there — not here):

- **Every** collider mesh must have its local `physics:approximation` cleared, so each runnable owns its own value (this is the per-collider form of the Neutral-base precondition above).
- The base carries the runtime-agnostic multibody topology (rigid bodies + `UsdPhysics` joints) and exactly one `PhysicsArticulationRootAPI`. If the topology or articulation root is missing, hand off to `FET_004_STANDARD`/`FET_004_NEWTON` before wiring the Newton variant.

## Worked Example: single-mesh prop

Add the `Newton` variant to a single-mesh prop in four edits (illustrated with a `RootNode`-rooted prop; substitute your asset's `defaultPrim` and collider mesh path). This is authored from the requirement docs — it does not depend on any existing asset.

1. Create the payload layer `runnables/physics/newton.usd` **inside the asset root**, in a `runnables/physics/` subfolder beside the root layer (e.g. `simready_usd/runnables/physics/newton.usd`), with the same `defaultPrim` as the root asset. Do not place `runnables/` outside the asset root — the `./`-anchored payload below must resolve within it (AA.001). Standardize on binary `.usd`; author the equivalent `.usda` below and export it to `.usd` (e.g. `Sdf.Layer.FindOrOpen("newton.usda").Export("newton.usd")`), or keep `.usda` if a binary cannot be compiled:

```usd
#usda 1.0
(
    defaultPrim = "RootNode"
    doc = """Newton overrides for sm_obs_orange_a02_01.usda"""
)

over "RootNode"
{
    over "Geometry"
    {
        over "obs_orange_01_obj_01"
        {
            over "obs_orange_01_mesh_01" (
                prepend apiSchemas = ["NewtonCollisionAPI", "PhysicsMeshCollisionAPI", "NewtonMeshCollisionAPI"]
            )
            {
                float newton:contactGap = 0
            }
        }
    }
}
```

2. In the root asset layer, add the `Newton` entry to `customLayerData.SimReady_Metadata.Variants.Physics` (alongside any `PhysX`/`MuJoCo` entries). The `Variants` dict lives **inside** `SimReady_Metadata`, never as a sibling top-level `customLayerData.Variants` key:

```usd
"Newton" = {
    string prim = "/RootNode"
    string variantSetName = "Newton"
    string activateOption = "Enabled"
}
```

3. On `RootNode`, add `"Newton"` to `prepend variantSets` and add its default selection to `variants` (`string Newton = "Disabled"`).

4. Add the `variantSet "Newton"` block with a payload-free `Disabled` option and an `Enabled` option that prepends the payload:

```usd
variantSet "Newton" = {
    "Disabled" {
    }
    "Enabled" (
        prepend payload = @./runnables/physics/newton.usd@
    ) {
    }
}
```

Note that the `Enabled` option holds only the payload arc and the `Disabled` option is empty (RV.010), and the payload authors only Newton + neutral data with no `Mjc*`/`mjc:*` (RV.011).

## Validation

The RV.004-RV.006 + RV.010/RV.011 checks live in `nv_core/tiers/simready_foundation_tier_core/simready/foundation/tier_core/capabilities/core/runtime_variants/validation.py` and are generated into `simready.foundation.tier_core.requirements` as `RV_001`-`RV_011`. To run them locally:

1. Build the core tier wheel, which is what generates the `RV_001`-`RV_011` enums from the capability markdown: `repo.bat build_tiers` (writes `nv_core/tiers/simready_foundation_tier_core/_build/dist/simready_foundation_tier_core-*.whl`).
2. Create a Python 3.12 venv: `repo.bat uv venv --python 3.12 .venv_val`.
3. Install USD, the validator deps, and the tier wheel into it: `repo.bat uv pip install --python .venv_val usd-core usd-validation-nvidia numpy nv_core/tiers/simready_foundation_tier_core/_build/dist/simready_foundation_tier_core-*.whl`.
4. Invoke the venv Python directly (not `uv run`, which spawns its own venv) against a small harness that imports `simready.foundation.tier_core.capabilities.core.runtime_variants.validation`, calls `ResetCaches()` then `CheckStage(stage)` on each checker, and reads results via `checker.GetIssues()`. Installing the wheel puts the generated enums on the path, so no `PYTHONPATH` setup is needed.

Confirm the Newton trio (`RV_004`/`RV_005`/`RV_006`) plus the isolation rules (`RV_010`/`RV_011`) pass on the `Disabled`-default stage. RV.011 recomposes each declared variant selection, so also spot-check that enabling Newton composes only Newton + neutral physics (no `mjc:*`, no `physics:approximation = "sdf"`). If the environment cannot be bootstrapped, fall back to inspecting the staged USD for the variant set, anchored payload, nested `SimReady_Metadata.Variants.Physics.Newton` entry, pure variant sections, and a Newton-only payload, and report validation as inspection-only.

## Limitations

- Do not silently mutate the source asset; work on the requested staged output.
- Do not author PhysX or MuJoCo variants in this skill.
- Keep runtime data out of the root variant sections (RV.010) and keep the Newton payload strictly Newton — no `Mjc*`/`mjc:*` or `Physx*`/`physx*:`, and no `physics:approximation = "sdf"` (RV.011).
- Do not create geometry; the single-mesh collision delta only annotates existing meshes.
- Beyond the deterministic single-mesh collision delta, do not invent Newton runtime physics values (joints, articulation/scene/material tuning) that require domain knowledge; scaffold and hand off to `FET_003_NEWTON`/`FET_004_NEWTON`/`FET_022_NEWTON`.

## Report Fields

| Field | Meaning |
|---|---|
| `feature` | Exact feature ID and version, for example `FET_000_NEWTON@0.1.0`. |
| `input` | Source asset or package inspected. |
| `output` | Staged output path, or `in-place` only when explicitly requested. |
| `requirements_repaired` | Requirement IDs repaired in this pass (subset of RV.004-RV.006, RV.010, RV.011). |
| `requirements_blocked` | Requirement IDs needing Newton override data or user input. |
| `validation` | Command or inspection used to verify this exact feature. |
| `status` | `passed`, `failed`, `skipped`, or `blocked`. |
| `next_step` | The next exact feature skill or evidence needed. |

From nvidia/simready-foundation at 60b8bdd, shown as plain text under its license: Apache-2.0 (permissive). The license text is in skills/simready-foundation-conform-fet-000-newton/SKILL.md.

View SKILL.md on GitHub at 60b8bdd

Files

3 files, 21 KB in total. Each file has a sha256 in the pin's manifest; the first 12 characters are shown.

Skill folder on GitHub at 60b8bdd
All 50 skills from NVIDIA

More from NVIDIA

  • error-explainer

    This skill helps the agent understand and explain compiler and runtime errors.

    Claude Code, Codex, Cursor +4 · about 66 tokens · pinned 7 days ago

  • order-lookup

    This skill retrieves and modifies individual orders in an e-commerce system.

    Claude Code, Codex, Cursor +4 · about 100 tokens · pinned 7 days ago

  • debug-openshell-cluster

    Diagnoses OpenShell gateway deployment problems across Docker, Podman, Kubernetes, and VM compute platforms.

    Claude Code, Codex, Cursor +4 · about 19.1K tokens · pinned 7 days ago

  • nemo-automodel-model-onboarding

    This skill guides adding or modifying model architecture support in NeMo AutoModel, covering discovery of model type from config, implementation of custom…

    Claude Code, Codex, Cursor +3 · about 5.6K tokens · pinned 7 days ago

  • nvflare-convert-lightning

    Converts existing PyTorch Lightning training code into an NVFLARE federated learning job by patching the trainer with the Lightning Client API, configuring…

    Contains scripts

    Claude Code, Codex, Cursor +3 · about 3K tokens · pinned 7 days ago

  • nvflare-convert-pytorch

    Converts existing plain PyTorch training code into NVFLARE federated learning jobs using the Client API for model exchange, local validation, and job export.

    Contains scripts

    Claude Code, Codex, Cursor +3 · about 2.5K tokens · pinned 7 days ago

All Coding workflow skills
  • planning

    This skill orchestrates multi-workflow projects with dependencies and shared data tables by creating a task graph in the orchestrator via the create-tasks tool.

    by n8nOfficial publisher

    Claude Code, Codex, Cursor +3 · about 2.4K tokens · pinned 7 days ago

  • progressive-building

    This skill implements incremental workflow development for agents building or extending workflows.

    by n8nOfficial publisher

    Claude Code, Codex, Cursor +3 · about 1.4K tokens · pinned 7 days ago

  • workflow-builder

    Builds and edits n8n workflows using the TypeScript SDK.

    by n8nOfficial publisher

    Claude Code, Codex, Cursor +3 · about 14.2K tokens · pinned 7 days ago

  • remotion-upgrade

    This skill upgrades Remotion and its related packages to the latest stable version.

    by RemotionOfficial publisher

    Claude Code, Codex, Cursor +3 · about 460 tokens · pinned 7 days ago

  • remotion-upgrade

    This skill upgrades Remotion and related packages across a project, automatically detecting the package manager and workspaces.

    by RemotionOfficial publisher

    Claude Code, Codex, Cursor +3 · about 460 tokens · pinned 7 days ago

  • final-release-review

    Audits Python SDK release candidates or plans against the previous release, comparing diffs for regressions, API changes, and version compatibility.

    by OpenAIOfficial publisher

    Contains scripts

    Claude Code, Codex, Cursor +3 · about 5.4K tokens · pinned 7 days ago

The skills listed here are third-party software, collected from public GitHub repositories as a catalog of recent skills and provided as is. Agent Native does not write or control them and is not responsible for any harm, loss or damage caused by installing or using them. Labels and checks describe what we saw at one commit and are not a guarantee. Read a skill's files before you run it. Terms of use