# atmos helm deploy

Deploy a Helm release for a component in a stack. `deploy` is an alias for
[`apply`](/cli/commands/helm/apply): it installs or upgrades the release
(`helm upgrade --install`), or delivers rendered manifests to a provision
target when `--target` is set.

## Usage

```shell
atmos helm deploy <component> --stack <stack> [options]
atmos helm deploy monitoring -s plat-ue2-dev --target deployment-repo

# Deploy components filtered by tags or labels (composes with --all/--affected)
atmos helm deploy --all --tags production,tier-1
atmos helm deploy --affected --labels cost-center=platform
```

## Flags

- **`--stack`, `-s` (required)**
  Atmos stack.
- **`--target` (optional)**
  Provision target to deliver to. Defaults to 
  `provision.default`
  , otherwise the cluster.
- **`--dependency-update` (optional)**
  Fetch declared chart dependencies when they are missing. See 
  [`apply`](/cli/commands/helm/apply#flags)
  .
- **`--all` / `--affected` / `--include-dependents` (optional)**
  Process multiple Helm components in dependency order. See 
  [`apply`](/cli/commands/helm/apply)
  .
- **`--tags` / `--labels` (optional)**
  Filter by tags (comma-separated, matches any) or labels (comma-separated 
  `key=value`
   or 
  `key:value`
   pairs, matches all): 
  `--tags=production,tier-1`
  , 
  `--labels=cost-center=platform`
  . Compose with 
  `--all`
  /
  `--affected`
   to narrow the selected set further; cannot be combined with a single component argument.
- **Release lifecycle flags (optional)**
  `--on-failure`
  , 
  `--cleanup-on-failure`
  , 
  `--wait`
  , 
  `--wait-for-jobs`
  , 
  `--timeout`
  , 
  `--history-max`
  , 
  `--no-hooks`
  , 
  `--skip-crds`
  , 
  `--server-side-apply`
  , and 
  `--force-conflicts`
   behave the same as for 
  [`apply`](/cli/commands/helm/apply#flags)
   and take precedence over stack 
  `release`
   configuration.