Launch Scene logo

Launch Scene

OrganizationPopular
AgibotTech
launch-scene

Launch a `genie_sim_bringup` scene against a chosen physics + render backend, using `ros2 launch genie_sim_bringup app.launch.py`. Covers the scene × launcher matrix (pick-and-place / whole-body-control / flat-table demos × Isaac PhysX / Newton-standalone backends) and the optional MoveIt 2 + WBC RViz overlay. Trigger: When the user asks to "launch a scene", "启动场景", "run pnp", "run wbc", "start the simulator", names any `scene_*.yaml` or `launcher_*.yaml`, or wants to bring up the RT Engine in interactive mode (with or without rviz / moveit).

Overview

PublisherAgibotTech
Repositorygenie_sim
Skill namelaunch-scene
Stars
1.4K
Forks
119
Bundled files
Instructions only
LicenseMPL-2.0
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by AgibotTech on GitHub. Read the source before you install it.

Installation

Install the Launch Scene AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/AgibotTech/genie_sim.git /tmp/genie_sim
mkdir -p .claude/skills
cp -r /tmp/genie_sim/source/geniesim_ros/skills/launch-scene .claude/skills/launch-scene
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Launch Scene in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Launch Scene on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Launch Scene is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

When to Use

  • User wants to bring up the Genie Sim RT Engine interactively — physics + render + ROS topics on one shared sim_time.
  • User references any scene yaml under source/geniesim_ros/src/ros_ws/src/genie_sim_bringup/config/scene_*.yaml or launcher yaml launcher_*.yaml.
  • User wants MoveIt 2 + WBC RViz on top of a running scene (Genie G2 only).

Do not use for:

  • Running a benchmark task → run-benchmark (under source/geniesim_benchmark/skills/).
  • Probing an inference server → check-inference.
  • Building the workspace first → build-workspace. Run that before this skill.

Critical Patterns

  1. Scene = robot + task; Launcher = physics engine + renderer. They're orthogonal axes — pair any scene with any launcher.
  2. Stable path is launcher_ovrtx_isaac_physx. Every launcher_newton_* and launcher_ovrtx_isaac_newton row is research / preview — physics behaviour, perf, and yaml schema can break between commits. Use the stable launcher unless the user explicitly wants the experimental rig (cloth / soft body / mjwarp).
  3. MoveIt is Genie G2 only. Other robots run on the engine but don't have a packaged MoveIt config.
  4. Inside the container. Every command below assumes geniesim docker into and source devel/setup.bash (see build-workspace skill).

The matrix

Scenes

SceneRobotWhat it showcases
scene_pnp_g2_opGenie G2 + omnipickerPick-and-place workflow
scene_wbc_g2_spGenie G2 + swiftpickerWhole-body control workflow
scene_flat_g2_spGenie G2 + swiftpicker, flat tableBaseline tabletop
scene_flat_aconeARX aconeReference robot (tabletop)
scene_flat_alohaAlohaReference robot (tabletop)
scene_flat_fr3Franka FR3Reference robot (tabletop)
scene_flat_ur5_robotiq_140 / _85UR5 + RobotiqReference robot (tabletop)

Launchers

LauncherPhysics engineRendererStatus
launcher_ovrtx_isaac_physxIsaac Sim PhysXStandalone OVRtx nodeStable — start here
launcher_ovrtx_isaac_newtonIsaac Sim Newton (wrapper)Standalone OVRtx node🧪 EXPERIMENTAL
launcher_newton_mjwarpNewton-standalone + mujoco-warpInline OVRtx✅ Stable (rigid only)
launcher_newton_fsvbdNewton-standalone + Featherstone + VBDInline OVRtx🧪 EXPERIMENTAL — cloth / soft body
launcher_newton_avbd / _mjvbd / _mjxpbdNewton-standalone, mixed solversInline OVRtx🧪 EXPERIMENTAL

Workflow

Step 1 — Collect inputs

Ask the user via AskUserQuestion if not given:

  • Scene: free-text basename (e.g. scene_pnp_g2_op). Auto-complete from the table above.
  • Launcher: default launcher_ovrtx_isaac_physx (stable). Only pick a launcher_newton_* if the user explicitly wants the experimental backend or cloth / soft body.
  • Headless? Default false (GUI window opens on $DISPLAY). Pick the value by where you're running:
    • Local workstation with a screenheadless:=false. The Isaac Sim viewport renders into a window you can see.
    • Remote / SSH / headless server / CIheadless:=true. No GUI, no $DISPLAY requirement. Physics + render still run; cameras publish over ROS exactly the same. Use this on machines without a display, when running over ssh without X-forwarding, or in container / batch jobs.
    • Remote machine, want the GUI? SSH with X-forwarding (ssh -X) or set $DISPLAY to a Xvfb / VNC server, then use headless:=false. Otherwise the launcher errors on the missing display rather than silently failing late.
  • MoveIt + RViz? Default false. Only meaningful for scene_*_g2_*.

Step 2 — Make sure the overlay is sourced

bash
source devel/setup.bash                   # see `build-workspace` skill if missing
ros2 pkg list | grep genie_sim_bringup    # must list the package

Step 3 — Launch the scene

bash
ros2 launch genie_sim_bringup app.launch.py \
  scene:=<SCENE> \
  launcher_config:=<LAUNCHER> \
  headless:=<true|false>

Recommended starting line for new users (P&P, stable physics, GUI):

bash
ros2 launch genie_sim_bringup app.launch.py \
  scene:=scene_pnp_g2_op \
  launcher_config:=launcher_ovrtx_isaac_physx \
  headless:=false

Step 4 — (optional) MoveIt 2 + WBC RViz

In a second container shell:

bash
source devel/setup.bash

# default Genie G2: crsB arm + swiftpicker gripper
ros2 launch genie_sim_moveit wbc.launch.py

# different arm/gripper combo (match whatever scene_*_g2_{op,sp} you launched):
ros2 launch genie_sim_moveit wbc.launch.py arm:=crs gripper:=omnipicker

arm:={crs,crsB} and gripper:={swiftpicker,omnipicker,none} are independent — the launch wires both into the URDF filename and the SRDF xacro mappings.

Step 5 — Verify topics

bash
ros2 topic list                            # /tf, /joint_states, /joint_command, /clock, …
ros2 topic hz /clock                       # should print sim-time pacing

Commands (copy-paste summary for the user)

bash
# Inside the container, after `geniesim ros build dev`:
source devel/setup.bash

# Pick-and-place + stable Isaac PhysX:
ros2 launch genie_sim_bringup app.launch.py \
  scene:=scene_pnp_g2_op \
  launcher_config:=launcher_ovrtx_isaac_physx \
  headless:=false   # local workstation with a screen; flip to true on a remote/headless host

# Whole-body control + Newton-standalone (experimental):
ros2 launch genie_sim_bringup app.launch.py \
  scene:=scene_wbc_g2_sp \
  launcher_config:=launcher_newton_mjwarp \
  headless:=false   # local workstation with a screen; flip to true on a remote/headless host

# Optional MoveIt 2 + WBC RViz (second shell):
ros2 launch genie_sim_moveit wbc.launch.py

Notes

  • The same app.launch.py covers every scene × launcher combo — there is no per-scene launch file. Don't try to invoke a yaml directly.
  • Newton GL viewer pose is configurable per scene: every scene_*.yaml carries a viewer_camera: block with pos: / lookat:.
  • For cloth / soft-body work, you need a launcher_newton_* that pairs with a VBD-family solver and a scene yaml that opts into the deformable. Treat any cloth / soft-body result as a preview, not a benchmark.
  • init_base_pose and init_joint_pos apply as a non-physics teleport on every backend — the robot spawns at the configured pose without an initial PD swing.

Resources

Frequently asked questions

What does the Launch Scene AI skill do?

Launch a `genie_sim_bringup` scene against a chosen physics + render backend, using `ros2 launch genie_sim_bringup app.launch.py`. Covers the scene × launcher matrix (pick-and-place / whole-body-control / flat-table demos × Isaac PhysX / Newton-standalone backends) and the optional MoveIt 2 + WBC RViz overlay. Trigger: When the user asks to "launch a scene", "启动场景", "run pnp", "run wbc", "start the simulator", names any `scene_*.yaml` or `launcher_*.yaml`, or wants to bring up the RT Engine in interactive mode (with or without rviz / moveit).

Why use Launch Scene on TypingMind?

Because you install it once and use it with any model. Launch Scene is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Launch Scene in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/AgibotTech/genie_sim/tree/main/source/geniesim_ros/skills/launch-scene. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Launch Scene?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Launch Scene?

As many as you like. As long as a model supports skills, you can use Launch Scene with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Launch Scene AI skill free?

Yes. It is published on GitHub by AgibotTech under the MPL-2.0 license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇