Automation

[stelbailey]

Description

Introduces the Automation app type to the platform - a new project type that lets users build and run multi-step pipelines composed of engine calls (database, model, vector, storage, function, app) and control modes (trigger, wait). This feature is currently hidden behind the admin functionality on the platform, if the user has admin ability and it is toggled on, they can create automation apps from the landing page or new app page and edit them through a new dedicated pipeline editor workspace.

Changes Made

• New app type: AUTOMATION project type added to NewAppModal (CreateProject(projectType=["AUTOMATION"]))
• App catalog: Admin escape hatch to surface AUTOMATION apps in the existing App catalog until a dedicated catalog
  entry exists
• AutomationWorkspace: Edit-mode workspace rendered when a user with edit access opens an automation app — hosts the
  pipeline form editor
• AutomationRenderer: Read-only step summary rendered for users without edit access (shows steps + last run status)
• Pipeline editor: Full node-based pipeline editor with forms for each node type:
  • Database Engine — SQL query or write against a registered RDBMS
  • Model Engine — LLM, embeddings, vision, NER
  • Vector Engine — search, add/delete documents, list
  • Storage Engine — list, download, upload, delete, read-base64
  • Function Engine — execute or streaming
  • App Engine — arbitrary Pixel expression with optional app context
  • Wait/Delay — configurable pause between steps
• Output preview: Per-node output preview with table, JSON, markdown, and vector-results render modes; output
  transform pipeline (raw, rows-as-objects, first-row, column, JSONPath)
• Dark mode SVG: Added Appautomation-dark.svg for the landing header dark variant
• File structure: All automation types, constants, and components colocated under automation-workspace/

How to Test

1. Log in as an admin user
2. From the App Catalog, click New App → select Automation → give it a name and create it
3. Open the app — you should land in the Automation editor (pipeline view)
4. Add a Trigger node, then add a Database Engine or Model Engine node and configure it
5. Click Run — verify the pipeline executes and per-step results appear in the output panel
6. Open the app as a non-editor user — verify the read-only Automation Renderer appears (step list + last run status badge)

Notes

• Depends on new backend reactors in Semoss

[snyk-io]

✅ Snyk checks have passed. No issues have been found so far.

Status	Scan Engine	Critical	High	Medium	Low	Total (0)
✅	Open Source Security	0	0	0	0	0 issues

💻 Catch issues earlier using the plugins for VS Code, JetBrains IDEs, Visual Studio, and Eclipse.