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

# CI-CD

TMS ONE integrates with **GitHub Actions** and **Azure DevOps Pipelines** so you can trigger automated Playwright test runs from your existing CI/CD pipeline — for example, as part of a deployment workflow.

***

## How test execution works in TMS ONE

Before setting up CI/CD, it helps to understand how TMS ONE actually runs Playwright tests.

When a test is executed — whether from the UI or from a pipeline — TMS ONE runs the Playwright scripts **on its own server infrastructure**. You do not need Playwright installed in your own environment or pipeline.

The execution flow works like this:

```text theme={null}
Trigger (UI or CI/CD pipeline)
        ↓
TMS ONE API receives the request
        ↓
Job is queued in a Redis-backed queue (Bull)
        ↓
A worker picks up the job and runs:
  → npx playwright test (JavaScript)
  → dotnet test (C#)
        ↓
HTML report generated, zipped, uploaded to S3
        ↓
Results saved to database
        ↓
Real-time status pushed to the UI via WebSocket
        ↓
Results visible in Execution History
```

**Key point:** The queue can handle up to 10 concurrent test runs. If multiple runs are triggered at once, they are queued and processed as workers become available. The API responds immediately when a run is submitted — the test execution happens in the background.

***

## How the CI/CD path works

The CI/CD integration works through a simple mechanism:

1. You authenticate TMS ONE with your GitHub or Azure account
2. TMS ONE connects to your repository
3. When you click **Run Workflow**, TMS ONE sends a `workflow_dispatch` event to your GitHub Actions workflow (or triggers an Azure Pipeline)
4. Your pipeline runs — and the workflow file you add to your repo contains a single step that **calls back to the TMS ONE API** to start the test run
5. TMS ONE queues and executes the tests on its server, exactly the same as a direct UI execution

The CI/CD pipeline itself does not run Playwright — it simply tells TMS ONE to start a test run. All execution happens on the TMS ONE server.

```text theme={null}
TMS ONE UI (Run Workflow)
        ↓
GitHub / Azure receives dispatch event
        ↓
Pipeline job executes:
  curl https://api.tmsone.app/api/v1/playwright/run-tests-ciCD?id=...&projectId=...&email=...
        ↓
TMS ONE API queues and runs the Playwright scripts on its server
        ↓
Results saved and visible in TMS ONE Execution History
```

This means your pipeline doesn't need any Playwright dependencies, test files, or browser setup. The workflow file is intentionally minimal.

***

## Setting up the integration

### Step 1 — Authenticate

Before connecting a repository, authenticate TMS ONE with your Git provider.

The **Authenticate** button appears in:

* The **API Runner** sidebar
* The **Test Builder → Automation** execute page (the **Execute Tests** view)

Click **Authenticate** and choose your provider:

**GitHub Authentication** — you are redirected to GitHub to authorise TMS ONE. Once authorised, you are returned to TMS ONE and your GitHub account is connected.

**Azure Authentication** — you are redirected to Microsoft login. Sign in with your Azure DevOps account. Once authorised, your Azure organisation is connected.

<Note>
  Authentication is per-user and persists across sessions. You only need to authenticate once. To switch accounts, authenticate again using the same dropdown.
</Note>

***

### Step 2 — Add the sample workflow file to your repository

From the **Run Workflow** modal, download the sample workflow file for your provider. Add it to your repository:

* **GitHub:** Place the file in `.github/workflows/` in your repository
* **Azure DevOps:** Add the YAML file to your pipeline configuration

The sample workflow contains a single step — a curl call to the TMS ONE run endpoint. Before committing, replace the placeholder values with your actual script IDs, project ID, and email address. These values identify which scripts to run and which account to attribute the run to.

<AccordionGroup>
  <Accordion title="GitHub Actions sample structure">
    The workflow is triggered via `workflow_dispatch` (a manual/API trigger). It runs on a standard `ubuntu-latest` runner and makes a single HTTP GET request to the TMS ONE API. No Playwright installation or test files are needed in the workflow.
  </Accordion>

  <Accordion title="Azure DevOps sample structure">
    The pipeline is triggered the same way. Use `ubuntu-latest` or another Microsoft-hosted agent — no special dependencies are required since the pipeline only makes an HTTP call to the TMS ONE API. If you have a self-hosted agent pool you prefer, that works too.
  </Accordion>
</AccordionGroup>

<Tip>
  After downloading the sample, replace the placeholder values (project ID, email) with your actual TMS ONE project details before committing the file to your repository.
</Tip>

***

### Step 3 — Connect repository & run

Once authenticated and your workflow file is in place, click **Run Workflow** on the **Test Builder → Automation** execute page.

Fill in the **Connect Repository & Run Workflow** modal:

**For GitHub:**

<ParamField path="Provider" type="select" required>
  Select **Connect GitHub Repository**.
</ParamField>

<ParamField path="GitHub Organisation" type="text">
  Auto-populated from your authenticated GitHub account. Read-only.
</ParamField>

<ParamField path="Repository" type="select" required>
  The repository where your workflow file lives.
</ParamField>

<ParamField path="Initial Branch" type="select" required>
  The branch to dispatch the workflow against.
</ParamField>

<ParamField path="Workflow ID" type="string" required>
  The filename of your GitHub Actions workflow file.\
  Example: `ci-workflow.yml`
</ParamField>

**For Azure DevOps:**

<ParamField path="Provider" type="select" required>
  Select **Connect Azure Repository**.
</ParamField>

<ParamField path="Azure Organisation / Project" type="text">
  Auto-populated from your authenticated Azure account. Read-only.
</ParamField>

<ParamField path="Repository" type="select" required>
  The Azure DevOps repository.
</ParamField>

<ParamField path="Initial Branch" type="select" required>
  The branch to run the pipeline against.
</ParamField>

<ParamField path="Workflow ID" type="string" required>
  The pipeline filename.\
  Example: `azure-pipeline.yml`
</ParamField>

Click **Run Workflow**. TMS ONE dispatches the event to your provider. Your pipeline starts, calls back to TMS ONE, and the test run is queued on the TMS ONE server.

***

## Viewing results

Results from CI/CD-triggered runs appear in [**Execution History**](/qa/execution-history) under the **Automation** tab — exactly the same as runs triggered directly from the UI. Each entry shows pass/fail counts, which scripts ran, and a link to download the full HTML report.

While a run is in progress, the execute page shows an **"Execution is in progress"** indicator. Real-time status updates are pushed to your browser automatically.

***

## CI/CD in the API Runner

In the **API Runner**, the same GitHub/Azure authentication also lets you **commit and push API collections** directly from TMS ONE to your connected repository. This keeps your API test collections version-controlled alongside your application code.

After authenticating, a **Commit & Push** option becomes available on your API collections.

***

## What's next?

<CardGroup cols={2}>
  <Card title="Test Builder" icon="wand-magic-sparkles" href="/qa/test-builder">
    Generate and manage Playwright automation scripts
  </Card>

  <Card title="Execution History" icon="chart-bar" href="/qa/execution-history">
    View results from all test runs
  </Card>
</CardGroup>