# Use Cases & Sample Prompts

Five things teams do with Salus through an assistant. Paste a prompt and adapt the names to your organization.

## Ship from your editor

Create a project, deploy it, and let the assistant watch the pipeline through to live.

```text
Deploy this repository to my staging environment. Watch the pipeline and
tell me when it's live — if the build fails, show me the failing step's
logs and explain what went wrong.
```

## Diagnose production

Pull logs, metrics, and revision history, and get an explanation instead of raw output.

```text
Pull the last hour of logs for checkout-api in production, group the
errors, and explain the most frequent one. Has it gotten worse since
the last deployment?
```

## Stay on top of security

Have vulnerability findings prioritized and explained against your actual dependencies.

```text
Check my production environment for vulnerabilities. Which ones are
critical, and which of our services do they actually affect? Suggest
what to upgrade first.
```

## Manage configuration

Work with environment variables across the hierarchy without clicking through settings.

```text
List the environment variables my staging deployment actually resolves,
and tell me which level each one comes from. Then set LOG_LEVEL to
debug on staging only.
```

## Operate infrastructure

Provision databases, roll back releases, or take a deployment offline — with confirmation before each step.

```text
The latest production release of checkout-api is misbehaving. Show me
its recent deployment revisions and roll back to the last known-good one.
```

Tip

**Read-only** access can diagnose but not act — give the assistant **Operator** or **Full** only where you want it acting. See the [MCP overview](https://salus.cloud/docs/mcp/).