Skip to content
Back to skills

Simulation Core

ASecurity

"RoboTwin simulator bootstrap, task-class API, robot/camera

  • 247 stars
  • 0 votes
  • 0 copies
  • 3 views
  • Added September 9, 2026
code-qualitygodebuggingapi

Works with

  • api

Security analysis

A100/100

Pro scans all 5 files and shows the line behind each finding

Scanned September 9, 2026

npx -y skills add VectorSpaceLab/AREX-Skill --skill simulation-core --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Simulation Core?

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

Security grade badge for Simulation Core
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/vectorspacelab-simulation-core-arex-skill/badge)](https://www.skillsdirectory.com/skills/vectorspacelab-simulation-core-arex-skill)

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: simulation-core
description: "RoboTwin simulator bootstrap, task-class API, robot/camera
  configuration, and render troubleshooting."
metadata:
  disco-role: operating
disable-model-invocation: true
license: MIT
---

# simulation-core

Use this skill for RoboTwin work that depends on the simulator core:
task classes, scene bootstrap, robot and camera wiring, embodiment config,
action helpers, render smoke checks, and simulation troubleshooting.
If you only have the generated skill tree, use the root [workspace bootstrapper](../../references/workspace-bootstrap.md) first to materialize a pinned public workspace before running simulator work.

## Owns
- `Base_Task` lifecycle and task subclass structure
- SAPIEN engine/scene/bootstrap and render refresh
- Robot, camera, and embodiment config interpretation
- Actor construction helpers and action helpers
- Render smoke and task-import smoke
- Simulation/debug troubleshooting for bootstrap, planning, and capture

## Route elsewhere
- Data download, normalization, and trajectory conversion → [`data-pipeline`](../data-pipeline/)
- Policy evaluation, XPolicyLab scheduling, and server setup → [`policy-eval`](../policy-eval/)

## Start here
1. Read [`references/api-reference.md`](references/api-reference.md) for the task and helper APIs.
2. Read [`references/configuration.md`](references/configuration.md) for task, camera, robot, and embodiment configs.
3. If you only need to confirm the render path, run [`scripts/check_render_smoke.py`](scripts/check_render_smoke.py).
4. If a task import fails before scene setup, check the asset prerequisites in [`references/troubleshooting.md`](references/troubleshooting.md).

## Native checks
- `render-smoke`: bundled SAPIEN bootstrap and camera capture smoke.
- `task-import-smoke`: import one representative task module after assets are present and verify its `Base_Task` path initializes cleanly.

## Good fits
- "How do I structure a new task class?"
- "Why does a grasp/place plan fail?"
- "Which config key controls cameras, embodiment, or eval step limits?"
- "Why does SAPIEN render bootstrap fail on this machine?"

## Not here
- Dataset download or HDF5 layout questions
- Policy-server, adapter, or multi-GPU eval questions
- LLM-based task generation or instruction writing

For deeper details, use the bundled references and keep simulator-specific debugging here.

Files in this skill

  • SKILL.md2.3 KB
  • references/api-reference.md7.8 KB
  • references/configuration.md5.2 KB
  • references/troubleshooting.md4 KB
  • scripts/check_render_smoke.py3.7 KB

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.

Comments

Loading comments…

Newsletter

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