simready-foundation-add-feature

작성자: nvidia

Use for adding SimReady feature docs, manifests, requirement mappings, validation strategy, and index entries.

npx skills add https://github.com/nvidia/simready-foundation --skill simready-foundation-add-feature

SimReady Add Feature

Purpose

Use this skill to add a brand-new SimReady feature under nv_core/sr_specs/docs/features/. A feature is a versioned runtime/use-case contract made from exact requirement IDs and optional feature dependencies.

Do not use this skill for a new version of an existing feature. Use simready-foundation-update-feature for that.

Prerequisites

Before editing, read:

  • AGENTS.md
  • nv_core/sr_specs/docs/guides/guides.md
  • nv_core/sr_specs/docs/guides/features/features.md
  • nv_core/sr_specs/docs/guides/naming_conventions.md
  • nv_core/sr_specs/docs/features/features.md
  • existing neighboring feature markdown and JSON files for the same domain

If the feature will be added to a profile, also read nv_core/sr_specs/docs/guides/profiles/profiles.md and the target profile markdown/TOML entries.

Inputs

Collect or infer:

InputRequirement
feature_numberNumeric ID such as 025. If absent, inspect existing feature IDs and propose the next appropriate number.
feature_idExact ID such as FET025_BASE_NEUTRAL; must match the variant pattern used by existing JSON manifests.
versionInitial semantic version, usually 0.1.0 unless the user states otherwise.
display_nameHuman-readable feature name.
runtime_promiseConcrete asset behavior the feature guarantees.
requirementsExisting requirement IDs or new requirement docs/validators needed by the feature.
dependenciesExact feature IDs and versions this feature depends on.
profile_targetsOptional profiles and versions that should adopt the feature.
validation_strategyAutomated validator, runtime test, manual test, or documented gap.
conform_skill_planNew conform skill name, existing conform skill to update, or documented reason no conform skill can safely repair the feature.

Instructions

Use this checklist when changing the repository:

  1. Inspect existing feature numbers, IDs, filenames, and variant suffixes. Avoid reusing an ID or inventing a new suffix when an existing suffix matches.
  2. Define the runtime promise in asset terms before choosing requirements.
  3. Prefer existing requirements and validators. Add new requirement docs or validators only when the feature needs a new testable rule.
  4. Create the JSON manifest under nv_core/sr_specs/docs/features/:
    • use the guide filename pattern FET_###_base_<variant>-<version>-<description>.json
    • include id, version, display_name, path, and requirements
    • include dependencies only when the feature actually depends on other features
    • use exact dependency versions
    • avoid circular dependencies
    • keep requirements explicit when feature expansion replaces a base requirement
  5. Create the feature markdown page under nv_core/sr_specs/docs/features/:
    • describe the runtime use case
    • list properties, dependencies, profiles, and requirements
    • link each requirement doc and validator implementation when available
    • document testing, samples, manual review, and known gaps
  6. Update nv_core/sr_specs/docs/features/features.md with the new feature row and toctree entry.
  7. Update nv_core/sr_specs/docs/features/feature-dependency-graph.md when the feature has dependencies or affects common dependency diagrams.
  8. Add a conform skill for the new feature when the feature can produce asset-level validation failures:
    • create skills/simready-foundation-conform-fet-###-<feature-name>/SKILL.md
    • include source-of-truth feature/requirement/validator paths
    • define what the skill may repair automatically and what it must block on
    • call out required model/tool capabilities, such as vision, CAD/source data, runtime simulation, or material identity
    • update assets/openai.yaml so the skill is discoverable
    • if the feature is metadata-only, advisory-only, or cannot be repaired safely, document the reason in the feature summary and validation handoff
  9. If profile adoption is requested, hand off to simready-foundation-add-profile or simready-foundation-update-profile; do not silently mutate an existing profile version.
  10. Validate consistency:
  • JSON parses
  • manifest id and version match the requested feature
  • every dependency feature/version exists
  • every requirement ID is documented or intentionally new in the same change
  • feature markdown and index links resolve by path/name
  • the matching conform skill exists or the no-conform-skill rationale is documented

Examples

Example request:

Add a new SimReady Foundation feature FET025 for factory connection points and wire it into feature docs and manifests.

Expected result summary:

changed_files: new docs, manifests, indexes, or validation scaffolding
validation: focused static checks and any relevant docs/build checks
remaining_gaps: requirement, validator, adapter, profile, or runtime-test follow-up

Policies

  • Feature versions are immutable once published or used by a profile.
  • Keep the feature focused on one runtime promise.
  • Do not include requirements just because they are nearby; include only what the feature needs.
  • If a technology-specific feature replaces a neutral/base requirement, list the full replacement requirement set explicitly instead of depending on the base feature for conflicting rules.
  • Treat profiles.toml as profile source of truth; feature docs should mention profile usage but not replace TOML.
  • Feature authoring and asset repair should evolve together. A new feature that introduces required authored USD data should normally ship with a conform skill for repairing or clearly blocking on that feature.

Limitations

  • Do not mutate published feature or profile versions in place.
  • Do not invent requirement IDs or validator behavior when the contract is ambiguous; record the question.
  • Do not skip index, manifest, validation, or downstream follow-up notes.

Troubleshooting

  • Error: the new concept overlaps an existing artifact. Solution: update the existing capability, requirement, feature, profile, or adapter instead.
  • Error: names or IDs conflict. Solution: re-check naming conventions and nearby indexes before editing further.
  • Error: validation strategy is unclear. Solution: document deferred validation and the exact follow-up skill.

Resources

  • assets/openai.yaml preserves optional UI metadata for clients that read skill display hints. It is not required for the workflow.

Summary Format

Report:

FieldMeaning
feature_idNew feature ID.
versionNew feature version.
feature_markdownFeature documentation path.
feature_manifestJSON manifest path.
requirementsRequirement IDs included.
dependenciesFeature dependencies included.
profiles_updatedProfiles changed, or none.
conform_skillNew/updated conform skill path, or documented rationale for none.
validationChecks run and remaining gaps.
next_stepProfile adoption, validator work, runtime test, or review.

nvidia의 다른 스킬

compileiq-debug
nvidia
Use when something is wrong: Search() hangs, all evaluations return INVALID_SCORE, scores aren't improving, every config returns the same number, ptxas errors…
official
create-github-pr
nvidia
gh CLI를 사용하여 GitHub 풀 리퀘스트를 생성합니다. 사용자가 새 PR을 만들거나, 코드 리뷰를 제출하거나, 풀 리퀘스트를 열고자 할 때 사용합니다. 트리거 키워드 -…
official
diagnose-perf
nvidia
First-responder performance triage for Isaac Sim and Isaac Lab. Identifies bottleneck category (GPU-bound, CPU-bound, VRAM, loading) using nvidia-smi and…
official
eagle3-review-logs
nvidia
Review EAGLE3 pipeline experiment logs from the launcher's experiments/ directory. Summarizes pass/fail status for all 4 tasks, diagnoses failures with root…
official
nemoclaw-maintainer-cross-issue-sweep
nvidia
다른 열린 이슈들을 스캔하여 주어진 PR이 함께 수정하거나 실수로 망가뜨릴 수 있는 이슈를 찾습니다. 인접 수정 기회와 모순 위험을 file:line…과 함께 출력합니다.
official
karpathy-guidelines
nvidia
일반적인 LLM 코딩 실수를 줄이기 위한 행동 지침입니다. 코드 작성, 검토 또는 리팩토링 시 과도한 복잡성을 피하고 정밀한 변경을 위해 사용하세요.
official
fhir-basics
nvidia
에이전트에게 FHIR R4 API의 작동 방식, 사용 가능한 리소스, 검색 매개변수를 사용한 쿼리 방법, 모든 응답 형식을 올바르게 파싱하는 방법을 가르칩니다…
official
underdeclared-agent
nvidia
A helpful assistant agent
official