▶️ EDIT

Sidebar

• Select loads the chosen project and exposes an Export button that writes <project>.zip to ${EXPORT_APPS} using the active export-app-filter. The rest of the page (expanders described below) becomes available from this view.

• Create and Clone lets you clone either the current project or one of the templates discovered on disk. The helper normalises the name (adds the _project suffix) and rejects duplicates before cloning.

• Rename copies the selected project to the new name, removes the original folder and switches the session to the renamed project once the operation succeeds.

• Delete permanently removes the active project after you tick the confirmation box. When the folder is deleted, AGILab automatically selects a remaining project so you can keep working.

• Import restores a project archive produced by Export. You can request a clean import (strips .gitignore files) and the modal overwrite flow guards against accidentally replacing an existing project.

Main Content Area

• When Select is active the page reveals a stack of expanders backed by the in-browser code editor. Each expander streams syntax-highlighted content and writes any saved changes straight back to disk:

  • PYTHON-ENV edits the project pyproject.toml so you can manage dependencies without leaving the browser.

  • PYTHON-ENV-EXTRA surfaces uv_config.toml for supplemental dependency constraints.

  • MANAGER and WORKER expose the main orchestration modules. The editor offers class/function/attribute pickers so you can focus on specific sections before saving.

  • EXPORT-APP-FILTER controls the .gitignore rules that the export flow applies when producing the archive.

  • APP-SETTINGS opens app_settings.toml. AGILab keeps the [args] and [pages] sections in sync with the Execute and Explore pages.

  • README allows quick edits to the project README.md.

  • APP-ARGS targets app_args.py in the app package. Use it to synchronise default arguments with the Execute page.

  • APP-ARGS-FORM displays (and creates if missing) the optional app_args_form.py Streamlit snippet used to render a custom parameter UI in Execute.

  • PRE-PROMPT serialises pre_prompt.json so you can tailor the prompt used by the Experiment assistant.

Support

Email: focus@thalesgroup.com