Digital Assets

The Digital Asset section is to store files on Rival. It’s a place to upload files - datasets, configuration, model weights, topic taxonomies, reference documents - and reference them as inputs from your tool code, without bundling the file into the tool itself.

You can find it at cortexone.rival.io/digital-assets.

Why it exists

Tool code on Rival is lean by design. Bundling a 200 MB dataset into your handler would slow cold starts and turn every code edit into a heavyweight redeploy. Digital Assets solves this by giving each file a stable, cloud-hosted path you can read at runtime - upload once, reference anywhere.

It’s also where you store topic files for Storm tools, configuration JSON for shared settings, or any large input your tool needs.

Asset properties

Each asset has three editable properties:

Name - a human-readable label
Description - optional notes about what the file contains or how it’s used
Visibility - Public or Private

Visibility

Public assets are accessible to anyone with the file path. Good for reference data, example inputs, sample taxonomies - anything you’re happy sharing.

Private assets are accessible only to tools owned by your organization. Use Private for proprietary datasets, credentials files, customer data, or anything that shouldn’t leave your workspace.

Size limit

Files and folders up to 500 MB are supported. Most common file types are supported, though videos aren’t allowed.

Using an asset in a tool

1. Upload the file

From /digital-assets, click Upload, select the file or folder, choose the category, give it a name, and pick visibility.

2. Copy the file path

Once uploaded, each asset has a stable path shown in its detail view (starting with mnt/). Copy it directly - don’t try to reconstruct it manually.

3. Reference the path in your handler

Read the path the same way you’d read any local file. The path format is identical across Python, JavaScript, and Lua runtimes.

# Example: reading a CSV asset in a Python tool
import csv

def cortexone_handler(event, context):
    with open("/assets/my-dataset.csv", "r") as f:
        reader = csv.DictReader(f)
        rows = list(reader)
    return {"statusCode": 200, "body": {"rows": rows}}

Managing assets

From the Digital Assets library you can:

Upload new files or folders
Edit the name, description, or visibility of an existing asset
Download an asset
Delete an asset

Only the Owner or Admin of the organization can delete any asset. Any public asset can’t be deleted.

See Digital Assets in building tools for the workflow inside the Tool Editor.