Read an AWS ARN, an Azure resource ID or a GCP resource name in one place, with the provider detected from the string's own shape.
Waiting for a resource identifier
An AWS ARN, an Azure resource ID or a GCP resource name — the provider is detected.
providers is a scope; after it, types and names alternate, so a nested resource has several pairs.The breakdown appears here
Each part with what it is for, and an explanation for any field that is legitimately empty.
The provider is detected from the string’s own shape, which is reliable because the three formats are unambiguous — an ARN starts arn:, an Azure ID starts /subscriptions/, and a GCP name is type-and-id pairs. Nothing here contacts a cloud provider: no credentials are needed or accepted, no API call is made, and the service and region data is maintained static reference from 2026-11 rather than anything live.
Cloud Resource ID Inspector runs entirely on your device — no cloud provider is contacted and no credentials are needed or accepted.
You’re on the Free plan. Compare plans
The provider is detected from the string itself, which matters because in practice you paste something out of a log or a Terraform state and do not especially want to tell a tool which cloud it came from. An ARN starts `arn:`; an Azure ID starts `/subscriptions/`; a GCP name is a run of type-and-id pairs. Those three are unambiguous, so the detection is reliable rather than a guess.
Azure IDs are the ones people get wrong, because the segment structure changes partway through. Everything up to `providers` is a scope — subscription, then resource group — and everything after it is the resource path within a provider namespace, where types and names alternate.
That means a nested resource, like a subnet inside a virtual network, has two type-and-name pairs after the provider. Reading it as a flat path gives you nonsense, and reading it as a fixed-length format gives you nonsense for anything nested.
A GCP resource name has no fixed length at all. It is pairs all the way down, so the sensible reading is "the last pair is the resource, everything before it is where it lives" — which is what gets reported.
Nothing is contacted. No cloud account, no credentials, no API call, and the service and region data is maintained static reference rather than anything live.
From the string's own shape, which is unambiguous across the three: an ARN begins `arn:`, an Azure ID begins `/subscriptions/`, and a GCP resource name is a sequence of type-and-id pairs. No guessing and no provider selector to get wrong.
Because it is nested. After the `providers` segment, Azure alternates resource types and names — so a subnet inside a virtual network has two pairs rather than one. That is exactly where a naive split goes wrong, and this reads the structure properly.
No. It reads the text you paste and nothing else. No credentials are needed or accepted, no API call is made, and nothing leaves your browser.