# Yam Simulation Transfer

> Use when borrowing LIBERO/robosuite strategy ideas for YAM grasping, handover, pouring, placement, rack, or bin tasks.

- Skill: `nvlabs/yam-simulation-transfer` (Agent Skill, multi-file: 8 files)
- Install (CLI): `npx skillmds@latest add nvlabs/yam-simulation-transfer`
- Raw SKILL.md: https://api.skillmd.com/api/skills/nvlabs/yam-simulation-transfer/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: nvlabs (https://skillmd.com/u/nvlabs)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/nvlabs/yam-simulation-transfer

---


# YAM Simulation Transfer

Borrow strategy, not simulator coordinates/API/constants.

The physical-robot implementation lives in `aspire/real`; the source
simulation workspace is its sibling `aspire/sim` (`../sim` when working from
the real workspace). Inspect simulation code and agent guidance there, then
convert the strategy into YAM-native behavior here.

References:

- `references/plate-libero-transfer.md`
- `references/bowl-libero-transfer.md`
- `references/can-libero-transfer.md`
- `references/libero-cup-mug-skills.md`
- `references/robosuite-two-arm-handover.md`
- `references/robosuite_two_arm_handover.py`
- `references/drawer-libero-transfer.md`

Convert ideas into YAM-native detections, frames, gripper geometry, planner
previews, and artifact checks before physical retries.