Folders for Sources and Automations: Tame the Sprawl

A workspace with five sources is a list you scan in a second. A workspace with eighty is a scroll. Once you are running ProxyHook across multiple clients, environments, and teams, the flat list stops being an index and starts being a chore: you scroll, you squint, you Ctrl-F the browser, and you still open the wrong automation half the time.

Folders fix that. You can now group your sources and automations into folders, nest them as deep as your structure needs, and read combined health totals for everything inside a folder without drilling in. Organization and observability in the same view.

The flat list does not scale

ProxyHook organizes work around three things: a source receives events, an automation routes them through optional filters and transformations, and a destination receives the result. That model is clean when you have a handful of each. It does not change shape as you grow, so a team running a hundred integrations sees the same single column a team running three does, just a hundred rows long.

The rows themselves are fine. The problem is that nothing relates them. Your Acme production billing webhook sits next to a staging Typeform source sits next to a client you offboarded last quarter, in whatever order they happened to be created. There is no grouping by client, by project, by environment, or by team, which is exactly how people actually think about their integrations.

Organize the way you already think

Folders live in both the Sources list and the Automations list. Create one, name it, and start filing items into it. The structure is yours to define:

• By client. An agency running ProxyHook for ten customers gets ten top-level folders, each a self-contained workspace.
• By environment. Separate Production from Staging so a test webhook is never mistaken for the real thing.
• By project or team. Group the sources and automations that belong to one initiative so the people who own it see only their slice.

Folders nest, so these axes combine. A real hierarchy might read Acme / Production / Billing, with the billing sources and automations for one client's production environment filed exactly where you would look for them. Go as shallow or as deep as your operation warrants.

One detail worth knowing up front: folders are scoped per area. A folder you create in Sources organizes sources only. Automations have their own folders. The two lists do not share a tree, because a source and an automation are different objects with different lifecycles, and forcing them into one hierarchy would mean every folder is half empty in both views.

Move items in bulk, not one at a time

Filing happens through multi-select. Check the sources or automations you want to group, choose Move Items to Folder, and they all move in one action. No dragging rows one at a time, no opening each item to reassign it.

The destination picker shows the full nested path of every folder, so when you are filing into a deep tree you pick Acme / Production directly instead of clicking down through Acme, then Production, then confirming. The path is the label.

The same picker includes a root option, written as /, that moves items back out of all folders to the top level. Reorganizing is symmetric: items go in through the picker and come back out through the same picker. If you nested too aggressively or a client structure changed, you are one bulk move away from flattening it.

Health at a glance, without drilling in

This is the part that makes ProxyHook folders more than a filing cabinet. Every folder row shows combined Success and Failure totals for the items directly inside it, aligned under the same columns as individual sources and automations. You do not open a folder to learn whether the integrations inside it are healthy. The row tells you.

Those totals are live, and they respect the date range you have selected. ProxyHook already gives you Activity analytics for every source and destination, a time-series breakdown of throughput, successes, and failures over a window you choose. Folder totals are that same accounting rolled up to the folder. Set the range to the last 24 hours and a folder row shows the last 24 hours of successes and failures across its items. Widen it to a week and the numbers widen with it.

What this buys you in practice: a Production folder showing a spike of failures is a triage signal you can read from the top-level list, before you have opened a single item. You scan the folder column the way you would scan a status board. When one folder's failure count is climbing, that is where you drill in, open the offending automation's Logs, and replay the events that failed once you have fixed the cause.

One precise note so the numbers never surprise you. The totals sum the items directly inside a folder for the selected range. They are direct-children totals, the successes and failures of the sources or automations filed at that level, not a deep recursive roll-up that reaches down through every sub-folder. Read a folder row as "the items in this folder," and the math always matches what you see when you drill in.

Drill in, and find your way back

Click a folder row to open it and view its contents. A breadcrumb shows where you are, for example Sources > Acme > Billing, and each segment is a jump-back link, so you can pop up one level or all the way to the top in a single click.

Two small behaviors keep navigation predictable. Clicking the sidebar nav for Sources or Automations always returns you to the top-level list, so the main nav is a reliable reset no matter how deep you have drilled. And the breadcrumb is always visible while you are inside a folder, so you never lose track of which slice of the workspace you are looking at.

Deleting a folder never deletes your data

Folders are organizational, not destructive, and the delete behavior reflects that. Rename and delete are inline actions that appear on hover over a folder row. When you delete a folder, the items inside it are un-filed back to the top level. They are not deleted.

This matters because the cost of reorganizing should be near zero. If deleting a folder risked taking a live source or a running automation with it, you would hesitate before ever restructuring, and the folders would ossify the moment you created them. Because a delete only removes the grouping, you can refactor your hierarchy as freely as your operation changes. Spin up a client folder, move things in, change your mind, delete it, and every source and automation is exactly where it was, back at the top level, still running.

Get started

If your sources and automations list has grown past the point of comfortable scanning, this is for you. Log in, create a folder in your Sources or Automations list, multi-select a few related items, and move them in. Then set your date range and watch the Success and Failure totals roll up per folder. A growing workspace stays an index instead of becoming a chore.