Compose nodes once, save the workflow, and run it on the next batch.
Manual runs are the shipped path. Upload, schedule, webhook, and change starts are planned in phases so the automation has the right org, credit, and safety guardrails.
Pick assets or a batch in the dashboard and run a saved workflow when you are ready.
Next phase: dashboard upload batches can start a workflow automatically after the files are ready.
Later phase: run saved workflows on a clock against saved collections, smart collections, or product groups.
Later phase: secure inbound webhooks can queue workflows after HMAC verification and safe asset import.
Product, asset, and approval change events need loop prevention before they become workflow triggers.
Drag actions onto the canvas and wire them together. Each action is a node, branch, approve, transform, write back, notify.
Background remove, upscale, generate variants, write alt text, detect objects, smart crop.
Route by attribute, file type, dimensions, brand, or any catalog field. If/else without code.
Send to a reviewer, hold for QA, or auto-approve when validation passes the spec floor.
Push to Shopify, your CDN folder, a vendor return path, or a webhook payload downstream.
Update product attributes, tags, status, or PIM fields when the run completes.
Email, Slack, or webhook on success, failure, or partial. One summary per batch, never per file.
Each template is a working pipeline you can clone, point at your products, and run from the dashboard. Triggered starts can attach later.
The current workflow engine runs saved pipelines on demand. The planned trigger layer will add upload batches first, then schedules and webhooks.
Workflows fail in messy ways when you build them yourself. The shipped runner handles execution history and retries. Trigger-specific budgets and runaway controls are planned before automatic starts ship.
Where to go next, depending on which side of the workflow you're building first.