Find and run requests across projects

Use the API catalog and search to find requests, assign who is responsible and run them in bulk.

In short

The API catalog lists every saved request in every environment across the projects you can see, with who is responsible and its last result. Filter it, assign a responsible person, and run many requests at once. Search anywhere finds requests by name or address.

Before you start

Who: anyone sees the requests of projects they can open. "Run selected" needs Run on each project, and a request that would run against Production at the Run level is left out.
Plan: every plan. Runs from the catalog count toward your allowance.
First: nothing to set up.

Steps

Open "Build" and choose "Catalog".
Search by name or URL, and narrow with the filters: project, environment, method, result, coverage, date and "My requests".
Read the "Coverage" column: a request has tests when it has checks of its own or is a step of a test suite - the rule the coverage figure and "Can we ship?" count by. Press "No tests" to see the requests nothing checks yet. "Saved views" keeps a set of filters under a name - "My failing requests" - to open again or share with your team; "All requests" clears it.
Tick rows and press "Run selected" to run them in their own project and environment, up to 100 at a time.
In the "Responsible" column, press "Assign" - or the name already there - and pick a member or a team. It applies to that environment only and gives no access.

Result: the runs are recorded in each project, and the catalog shows the new results.

Good to know

The filters live in the page address, so you can bookmark or share a filtered view.
Each row shows who is responsible, how many test cases and runs it has, and its last result.
Search anywhere (Ctrl+K, or ⌘K on a Mac) finds projects, saved requests by name or address, and test suites by name, limited to what you can reach, and forgives a typo when nothing matches exactly. Choosing a request opens it over the current screen; "Send" runs it with its saved headers, inherited credentials and the environment the project last ran in. "Open in API Tester" takes you to it to edit.
Below 768 px the table becomes labelled cards. The first filters sit in the row, the rest behind "More filters", and on a phone all of them are in one sheet. Every filter that is on shows as a chip you can remove, and one press clears them all.

If something goes wrong

A request is missing: it is in a project you cannot open.
Some selected rows were not run: you may only view their project, or they would run against Production at the Run level.