Find every stale variable across your Postman workspaces
One pass reads thirty-one variables and fourteen requests with their scripts, tests and auth, then names the four still pointing at a host retired in June.
The staging environment has thirty-one variables. Four are stale, one points at a host decommissioned in June, and nobody has cleaned it up because finding out which four means opening the environment, opening the collection, and reading both against a list of live services that exists only in somebody’s head.
A companion reads every workspace, every collection in full, and every environment, then hands you the four names and where each one is used. Then it files the ticket, messages whoever owns the retired host, and writes the row into the sheet. Strawberry connects to Postman through a native integration with one API key.
Check every Postman variable in one pass
Fetching a collection returns the entire document: thirty-one variables and fourteen requests with their pre-request scripts, tests and auth configuration attached. That answers what people open Postman for. Which requests still point at the retired host. Which environments define a variable no request references. Which of the four auth setups drifted, and when.
Because the answer is a document rather than a screenshot, it can go somewhere: a ticket, a review comment, or a sheet listing every workspace with what is stale inside it. That is data extraction over a tool most people never thought to query, and auditing a team’s estate is the kind of operations chore that stays undone because it is nobody’s deadline.
What executes here, and what gets read?
One thing executes: triggering a monitor, which runs whatever that monitor was configured to run, on Postman’s infrastructure, against whatever it points at. It calls real external services, spends your monitor run quota, defaults to asynchronous, and waits for your approval first, after which the companion fetches the monitor to read the run state.
Everything else is reading, and reading is where this earns its place: a request’s method, URL, headers, body, auth and tests, across every collection in every workspace. For most people Postman means send a request, so it is worth being exact that the runner stays in the app and the estate-wide audit is what happens here.
How careful is the variable merge?
More careful than a blind write, which matters because variables carry hosts and secrets into requests. The collection variable tool reads the remote collection first and then changes only matching keys or appends new ones. It does not replace the collection, and the environment variable tool treats the current remote environment as its source of truth in the same way.
Both merge scalar variables only. Neither rewrites requests, scripts, tests or auth, so a base URL update stays a base URL update. That boundary is what makes this safe to point at a shared workspace, and the calls still ask before they run.
- Read: workspaces, collections in full, environments, API specs, mocks, monitors.
- Write: create a collection, merge collection variables, merge environment variables.
- Execute: trigger a monitor run (asynchronous, quota-consuming, approval-gated).
- Absent: any request or collection runner, and any delete tool at all.
File the ticket the audit produced
The ticket naming the four stale variables.
The sheet listing every workspace with what is wrong inside it. The message to whoever owns the one pointing at a decommissioned host. A companion files all three in the same session, which is the half that never happens when the audit ends in a chat window.
Nothing deletes, which is a reassuring absence on a shared workspace: no collection, environment, mock, monitor or spec can be removed here, environments are not creatable, and API specs are readable rather than editable. Postman’s own AI sits beside the runner with a request in front of it, as of this writing.
Experience Strawberry for free
DownloadTrusted by fast-growing companies worldwide
Frequently asked questions
Strawberry is free to download and includes AI credits to start. Paid plans begin at $20/month. See pricing. · Reviewed · Canonical facts for AI agents