Part 5.2: Streamlit Integration II

Deploy the tutorial fixed-income dashboard with the CLI

Introduction

In Part 5.1, you built the tutorial dashboard under:

dashboards/tutorial_fixed_income_dashboard/

In this chapter, you will deploy that dashboard using the current CLI workflow.

The deployment flow is:

1. sync the project so the dashboard files are committed and pushed
2. create a project image for that pushed commit
3. list the project resources discovered from that commit
4. create a Streamlit dashboard release from the dashboard app.py
5. verify the resulting release

1) Sync the project

From the project root:

cd /path/to/your/project
mainsequence project sync -m "Add tutorial fixed-income dashboard"

This is the easiest way to make sure the dashboard files are:

• committed
• pushed
• visible to the backend resource discovery flow

2) Create a project image

List existing images first:

mainsequence project images list

If no image exists for the pushed commit, create one:

mainsequence project images create

The image must be built from a pushed commit. The CLI will either prompt you to select a pushed commit or let you pass one explicitly.

3) List project resources

Once the commit is pushed, list the resources discovered from the current remote head:

mainsequence project project_resource list

Find the resource ids for:

• dashboards/tutorial_fixed_income_dashboard/app.py
• dashboards/tutorial_fixed_income_dashboard/README.md

You will also need the image id from the previous step.

4) Create the dashboard release

Create the dashboard release with:

mainsequence project project_resource create_dashboard \
  --related-image-id <IMAGE_ID> \
  --resource-id <APP_RESOURCE_ID> \
  --readme-resource-id <README_RESOURCE_ID>

This is the current CLI flow for dashboard deployment.

Behind the scenes:

• the selected image defines the runtime environment
• the selected resource defines which app.py is released
• the CLI filters eligible resources so they match the same repo commit as the selected image

5) Verify the deployment

The simplest verification flow is:

1. confirm the image exists with mainsequence project images list
2. confirm the dashboard resources exist with mainsequence project project_resource list
3. confirm create_dashboard returned a release id

If you want a direct SDK verification, you can query releases for the dashboard resource:

PYTHONPATH=/Users/jose/code/MainSequenceClientSide/mainsequence-sdk \
.venv/bin/python - <<'PY'
from mainsequence.client.models_helpers import ProjectResource, ResourceRelease

app_resource = ProjectResource.get(path="dashboards/tutorial_fixed_income_dashboard/app.py")
releases = ResourceRelease.filter(resource=app_resource.id)

for release in releases:
    print(
        {
            "id": release.id,
            "release_kind": release.release_kind,
            "subdomain": release.subdomain,
            "related_image": release.related_image,
        }
    )
PY

If the release exists and points to the image you selected, the deployment path is correct.

6) Why this deployment chapter is different from the older draft

This version is intentionally narrower and more accurate:

• it does not ask you to clone an external dashboard repository
• it does not rely on a UI-first deployment flow
• it uses the current image/resource/release CLI commands
• it keeps dashboard construction and dashboard deployment as separate steps

7) Troubleshooting

• project_resource list shows no dashboard files: The dashboard files are not part of the pushed commit the backend is inspecting yet.
• images list shows no image for the commit: Create one with mainsequence project images create.
• create_dashboard says no matching resources exist: The selected image commit and the selected resource commit do not match.
• The dashboard opens but shows no rows: Use the landing-page refresh button to rebuild the tutorial assets and simulated prices.

At this point, your tutorial project has a current Streamlit dashboard and a current CLI deployment path that matches the SDK behavior in use today.

For the reusable helper layer behind these dashboard chapters, see Streamlit Helpers.