Skip to content
Back to skills

Simready Foundation Conform Fet 008 Object Forward

ASecurity

Author explicit local Object Forward from a caller-provided vector or visual evidence, then validate the sole requirement AF.001 on staged output.

  • 198 stars
  • 0 votes
  • 0 copies
  • 1 view
  • Added November 20, 2026
databases

Security analysis

A100/100

Pro scans all 2 files and shows the line behind each finding

Scanned November 20, 2026

npx -y skills add NVIDIA/simready-foundation --skill simready-foundation-conform-fet-008-object-forward --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Simready Foundation Conform Fet 008 Object Forward?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Simready Foundation Conform Fet 008 Object Forward
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/nvidia-simready-foundation-conform-fet-008-object-forward/badge)](https://www.skillsdirectory.com/skills/nvidia-simready-foundation-conform-fet-008-object-forward)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: simready-foundation-conform-fet-008-object-forward
description: Author explicit local Object Forward from a caller-provided vector or visual evidence, then validate the sole requirement AF.001 on staged output.
license: Apache-2.0
metadata:
  author: "Shaad Boochoon <sboochoon@nvidia.com>"
  tags:
    - simready
    - conformance
    - object-forward
---

# Conform Object Forward

## OpenUSD Competence

Running this conform skill is expert USD authoring. Reason from the OpenUSD /
Pixar USD spec rather than pattern-matching a validator message: composition
arcs and LIVRPS strength ordering, value resolution and list editing
(prepend, append, explicit, delete), the `def`/`over`/`class`
specifiers, native instancing and model hierarchy, stage units and up-axis,
and the `UsdGeom`, `UsdShade`, and `UsdPhysics` schemas. Make the exact native
edit on the correct prim, arc, and layer, derive values from the asset's real
geometry and composition, then reopen the composed stage to confirm the
opinion resolves as intended before claiming the gate passes. Never fabricate
data to satisfy a validator token. The repository-root `AGENTS.md`
"Master-Level OpenUSD Competence for Conform Work" section is the full
statement.

Targets `FET008_ObjectForward.0-1-0`, which owns AF.001 directly.
Read the feature and AF.001 docs first. Preserve source assets and reports.

1. Copy the asset and required dependencies into the requested output directory.
2. Select the defaultPrim or explicitly requested component; inspect geometry,
   renders or viewport evidence to identify its semantic front. Record the
   rationale and mapping into prim-local coordinates. If neither the caller
   nor evidence identifies a front, ask for intent instead of guessing.
3. Supply a finite unit vector to
   `simready.foundation.tier_core.asset_frame.author_forward_direction` on the
   staged prim. Repair malformed existing properties explicitly against AF.001
   before calling; the helper intentionally rejects conflicting authored data.
4. Preserve native upAxis. Do not author a custom upDirection or replace GR.002.
   Components author their own facing; there is no namespace inheritance.
5. Save staged layers and rerun the narrow AF.001 gate. Preserve its report.
6. When runtime evidence is requested, run object_forward_contract and
   object_forward_camera_response through simready-evaluate in Kit/Isaac to
   query AF.001 data and measure FET008 camera consumption on the loaded stage. Preserve metrics/captures and report runtime unavailability honestly.
   Profile conformance requires the other gates too.

Automatic migration from an asset with no semantic front is not defined.

Files in this skill

  • SKILL.md2.6 KB
  • assets/openai.yaml242 B

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

View all in databases →

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.