> ## Documentation Index
> Fetch the complete documentation index at: https://docs.testdino.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Run selected Playwright tests from Test Runs

> Pick Playwright tests from your test runs or new spec files on GitHub, and start a GitHub Actions run of only those tests.

**Run tests** starts a GitHub Actions run of only the Playwright tests you pick from the Test Runs page. You pick tests TestDino has already seen, or spec files on GitHub it has never seen, then choose a branch, a workflow, and the browsers. The results arrive as a new test run in [Test Runs](https://docs.testdino.com/platform/playwright-test-runs).

> [!NOTE]
>
> Before you start, you need:
>
> * The Owner, Admin, or Member role. Every plan includes **Run tests**.
> * GitHub connected to the project. If it is not, the dialog asks you to connect it first. See [GitHub App](https://docs.testdino.com/integrations/ci-cd/github.md).
> * A workflow with the `testdino_selection` input, as shown below.
> * `@testdino/playwright` 2.11.1 or later and Playwright 1.56 or later on the branch you run. Running whole spec files from **New on GitHub** needs `@testdino/playwright` 2.11.6 or later.

## Set up your workflow

**Run tests** uses the same `testdino_selection` workflow input as [Run automation](https://docs.testdino.com/test-management/run-automated-tests.md) in Test Management. A workflow already set up for Run automation needs no change.

Otherwise, add the input to the GitHub Actions workflow that runs your Playwright tests, and pass it to `npx tdpw test`. Follow [Set up your workflow](https://docs.testdino.com/test-management/run-automated-tests.md#set-up-your-workflow) for the lines to add and the rules for the test command. Add the workflow to every branch you want to run from TestDino.

## When the Run tests button appears

**Run tests** sits in the Test Runs page header, next to the **Sync** icon. It appears for the Owner, Admin, and Member roles when the project has test runs in the last 90 days and 1 of these is true:

| Condition | Example |
| :- | :- |
| GitHub is connected to the project | Runs come from any CI |
| Runs come from GitHub Actions | GitHub is not connected yet; the dialog asks you to connect it |
| Every run is a local run | Tests run from developer machines only |

The button is hidden for the Viewer and Billing Manager roles, and when runs come only from another CI provider and GitHub is not connected.

## Start a run

**1. Open Run tests**

On the Test Runs page, click **Run tests**. The dialog opens on **Pick tests** with every test TestDino has seen, and no filter applied.

**2. Pick tests**

Tick tests, or tick a folder, spec file, or describe block to pick every test under it. Click a folder, spec file, or describe block row to expand or collapse it; click a test row to tick it. Switch between the tree view (folder, spec file, describe block, test) and the list view with the 2 buttons next to the search box. **All** shows what the search and filters leave, **Selected** shows your picks. **Select all** picks every test shown; **Clear** removes your picks.

| Filter | Shows |
| :- | :- |
| **Status** | **Failed in latest run**, **Failed in last 30 days**, or **Flaky in last 30 days** |
| **Branch** | Tests that ran on that branch. **Failed in latest run** then uses that branch's latest test run. |
| **Browser** | Tests that ran in that browser |
| **Tag** | Tests with any of the chosen tags |
| **Folder** | Tests in any of the chosen folders. Hidden when your tests are not in folders. |
| **From a run** | The tests of 1 finished test run |

The list shows only tests from the repository connected to the project, because that is where the run starts. If the project also has test runs from other repositories, a note above the list names them.

Tests not reported in the last 30 days sit in a **Not run in 30+ days** group at the bottom.

A run takes up to 10,000 tests and 500 new spec files. Click **Next**.

**3. Choose a branch, browsers, and workflow**

* **Branch**: starts as the branch you filtered on in step 1, otherwise the repository's default branch. Search the list or type a name.
* **Browsers**: leave it on **All browsers** to run each test in the browsers it recently ran in, or pick 1 or more browsers. The list shows how many of your picks have run in each browser.
* **Workflow**: the workflows on that branch that are set up for TestDino. `N workflows can't run selected tests. Show why` lists the others and what to change.
* **If a test isn't on `<branch>`**: **Skip missing tests** (the default) or **Fail the job**. See [Tests missing on the branch](https://docs.testdino.com/test-management/run-automated-tests.md#tests-missing-on-the-branch).

**Tests to run** shows how many of your picks will run, such as `42 of 47 tests will run`, and a line for each reason a pick is skipped. Click it to list the tests that will run, and click **Show tests** on a reason to list the tests it skips. See [Tests skipped before the run](#tests-skipped-before-the-run).

**4. Run**

Click the **Run** button, such as `Run 42 tests`. The dialog shows **Run started**. Click **View in GitHub** to open the workflow run, or **Done** to close the dialog.

Closing the dialog after you pick tests or change a setting asks `Discard your test picks?`. Click **Keep editing** to return to your picks.

## Run spec files that are new on GitHub

**New on GitHub** lists spec files on the branch that TestDino has never seen, such as a test file added since the last test run. Picking one runs every test in it. Files with no test in them, such as a fully commented-out file, are left out.

The group appears below the tests in the **All** view, without a **Status**, **Tag**, or **From a run** filter. While TestDino checks the branch, the list shows `Checking GitHub for new spec files…`. It appears only when the latest test run used `@testdino/playwright` 2.11.6 or later.

## See the results

The results arrive as a new test run in [Test Runs](https://docs.testdino.com/platform/playwright-test-runs) when the GitHub run finishes. It shows who started it, such as `Manual · <name>`. To retry only the tests that failed, open the new test run and click **Re-run**. See [Re-run failed tests](https://docs.testdino.com/guides/debug-playwright-failures/rerun-failed-tests.md).

## Tests skipped before the run

**Tests to run** lists each reason a pick won't run, before CI starts. Each line starts with what won't run, such as `2 tests` or `1 new file`:

| Line | What to do |
| :- | :- |
| `… haven't run on <browser> recently` | The tests have no recent result in the browsers you chose. Clear the browser choice so it reads **All browsers**, or pick another browser. |
| `… have no browser to run in` | The tests haven't run in any browser in the last 90 days. Run your suite once, then try again. |
| `… can't be targeted in CI` | The test title has a `›`, a line break, or extra spaces, or is too long. Rename the test. |
| `… are no longer in TestDino` | TestDino has no record of these tests. If they still exist, run your suite once so TestDino records them. |
| `… aren't on <branch>` | The spec file, or the new file, is not on the chosen branch. With **Skip missing tests** they are skipped; with **Fail the job** the other tests run, then the CI job fails. |

When no pick can run, the card shows `None of these tests will run` and the **Run tests** button stays disabled.

## Troubleshooting

> [accordion] No workflow on \<branch> can run selected tests
>
> No active workflow on the chosen branch has the `testdino_selection` input. The dialog shows **Set up a workflow** with the lines to add. Add them on that branch, then come back and try again. **Show why** explains the other workflows.

> [accordion] GitHub has no branch named \<branch>.
>
> The branch is not on GitHub, so nothing can run and **Run tests** stays disabled. Pick an existing branch, or push the branch first.

> [accordion] New on GitHub doesn't appear
>
> The group shows only when the latest test run used `@testdino/playwright` 2.11.6 or later, in the **All** view without a **Status**, **Tag**, or **From a run** filter. Upgrade the CLI and run your tests once, or clear the filter. A branch whose spec files TestDino has all seen shows no group.

> [accordion] Selection \<id> includes whole test files, which need a newer @testdino/playwright
>
> The workflow ran a selection with a file from **New on GitHub** on `@testdino/playwright` older than 2.11.6, so nothing ran. Upgrade it on that branch, then start the run again.

> [accordion] A test is skipped as not run on \<browser>
>
> The test hasn't run recently in the browser you chose. Clear the browser choice so **Browsers** reads **All browsers**, and the test runs where it last ran.

> [accordion] Some tests are missing from the list
>
> The list shows only tests from the repository connected to the project. Tests reported from another repository are not listed, and the note above the list names those repositories. To run them, send that repository's test runs to a project connected to it, and run them from that project.

> [accordion] The Run tests button is missing
>
> The button needs the Owner, Admin, or Member role and test runs in the last 90 days. With runs only from another CI provider, connect GitHub to the project first. See [When the Run tests button appears](#when-the-run-tests-button-appears).

## Related

> [card] [Run automated tests](https://docs.testdino.com/test-management/run-automated-tests.md)
>
> Run the automated tests linked to test cases in Test Management.

> [card] [Re-run failed tests](https://docs.testdino.com/guides/debug-playwright-failures/rerun-failed-tests.md)
>
> Run only the failed or flaky tests from a finished test run.

> [card] [Node.js CLI](https://docs.testdino.com/cli/testdino-playwright-nodejs)
>
> Every `npx tdpw test` flag, including `--selection`.

> [card] [GitHub App](https://docs.testdino.com/integrations/ci-cd/github.md)
>
> Connect GitHub to a TestDino project.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.