A project is the unit of work in Prometheux. It groups everything you build around a problem — concepts, ontologies, data source connections, and results — so you can manage, share, and move that work as a whole.

​
Project scopes

Every project belongs to a scope that controls who can see and use it:
  • user — private to your own workspace (the default scope).
  • organization — shared across your organization.
When listing projects you can filter by one or more scopes, and when copying or importing a project you choose the scope it lands in.

​
Working with projects

The platform supports the full lifecycle of a project. Each of these operations is also available through the SDK and REST API (see the Projects API reference):
  • Create or update — save a new project with a name and optional description, or update an existing one.
  • List — browse the projects available in your selected scopes.
  • Load — open a specific project by its ID.
  • Copy — duplicate a project within or between scopes/workspaces.
  • Export / Import — export a project (including its data and metadata) to a portable file, and import it back into a workspace later.
  • Cleanup — delete a single project or remove all of your project resources.

​
Templates

Prometheux ships with a marketplace of project templates — pre-built starting points for common use cases (for example, financial analysis or customer analytics). You can browse the available templates and import one as a new project to get going quickly without building from scratch.

​
Workspaces

A workspace holds all the projects for a given scope. You can export an entire workspace — every project and its associated data — and import it back elsewhere, which is useful for backups or moving work between environments.

​
Quick example (SDK)

The snippet below shows the core project lifecycle using the Python SDK: 
import prometheux_chain as px
import os

# Set up authentication and configuration
os.environ['PMTX_TOKEN'] = 'my_pmtx_token'
px.config.set('JARVISPY_URL', "https://api.prometheux.ai/jarvispy/my-org/my-user")

# Create a new project
project_id = px.save_project(project_name="test_project")

# List all projects
projects = px.list_projects()
print(f"Available projects: {projects}")

# Load the created project
project_data = px.load_project(project_id)

# Clean up when done
px.cleanup_projects(project_id=project_id)

​
Next steps