Create new project or task

Create new project or task#

The template generator creates external Isaac Lab projects and internal tasks.

The template generator enables you to create an:

  • External project (recommended): An independent repository that can also load as an Isaac Sim extension.

    Hint

    The generator initializes a Git repository in the selected directory.

  • Internal task: A task added directly to the Isaac Lab repository for contribution upstream.

    Warning

    Pip installations of Isaac Lab do not support Internal templates. If isaaclab is loaded from site-packages or dist-packages, the Internal option is disabled and the External template will be used instead.

Running the template generator#

Install Isaac Lab by following the installation guide. Then run the generator from the uv-managed environment:

uv run isaaclab --new  # or "uv run isaaclab -n"

The generator will guide you in setting up the project/task for your needs by asking you the following questions:

  • Type of project/task (external or internal), and project/task path or names according to the selected type.

  • Isaac Lab workflows (see Task Design Workflows).

  • Reinforcement learning libraries (see Reinforcement Learning Library Comparison), and algorithms (if the selected libraries support multiple algorithms).

External project usage (once generated)#

Once the external project is generated, a README.md file will be created in the specified directory. This file will contain instructions on how to install the project and run the tasks.

The generated project is a uv workspace. From its root, create the environment and install the generated package in editable mode:

uv sync

  • List the tasks and physics presets available in the project.

    uv run python scripts/list_envs.py --show_presets

  • Train and play a task with the installed Isaac Lab commands.

    uv run isaaclab train --rl_library <library> --task <Task-Name>
uv run isaaclab play --rl_library <library> --task <Task-Name> --checkpoint latest

    The same command surface provides zero_agent, random_agent, benchmark, and train_multigpu. Generated packages advertise their task modules through package metadata, so the commands discover downstream tasks without project-specific runner scripts.

For more details, please follow the instructions in the generated project’s README.md file.

Internal task usage (once generated)#

Once the internal task is generated, it will be available along with the rest of the Isaac Lab tasks.

  • List the tasks available in Isaac Lab.

    uv run python scripts/environments/list_envs.py --show_presets

  • Run a task.

    uv run isaaclab train --rl_library <library> --task <Task-Name>

  • Run a task with dummy agents.

    Dummy agents help verify the environment before training.

    • Zero-action agent

      uv run isaaclab zero_agent --task <Task-Name>

    • Random-action agent

      uv run isaaclab random_agent --task <Task-Name>