| `artifactory.username`      | Username of the Artifactory user with read permissions that the agent will use. |
| `artifactory.password`      | Password or API token for the Artifactory user.                                 |
| `artifactory.image_filters` | Optional: List of repositories and images to analyze.                           |

If you don't specify any repositories in `artifactory.image_filters`, the agent
runs image analysis on all images in your Artifactory instance.

The following snippet shows a sample configuration:

```json
{
  "agent_id": "acme-prod-agent",
  "docker": {
    "organization_name": "acme",
    "username": "mobythewhale",
    "pat": "dckr_pat__dsaCAs_xL3kNyupAa7dwO1alwg"
  },
  "artifactory": [
    {
      "base_url": "https://acme.jfrog.io",
      "username": "acmeagent",
      "password": "hayKMvFKkFp42RAwKz2K",
      "image_filters": [
        {
          "repository": "dev-local",
          "images": ["internal/repo1", "internal/repo2"]
        },
        {
          "repository": "prod-local",
          "images": ["staging/repo1", "prod/repo1"]
        }
      ]
    }
  ]
}
```

Create a configuration file and save it somewhere on the server where you plan
to run the agent. For example, `/var/opt/artifactory-agent/config.json`.

#### Run the agent

The following example shows how to run the Docker Scout Artifactory agent using
`docker run`. This command creates a bind mount for the directory containing the
JSON configuration file created earlier at `/opt/artifactory-agent/data` inside
the container. Make sure the mount path you use is the directory containing the
`config.json` file.

<!-- prettier-ignore -->
> [!IMPORTANT]
>
> Use the `v1` tag of the Artifactory agent image. Don't use the `latest` tag as
> doing so may incur breaking changes.

```console
$ docker run \
  --mount type=bind,src=/var/opt/artifactory-agent,target=/opt/artifactory-agent/data \
  docker/artifactory-agent:v1
```

#### Analyzing pre-existing data

By default the agent detects and analyzes images as they're created and
updated. If you want to use the agent to analyze pre-existing images, you
can use backfill mode. Use the `--backfill-from=TIME` command line option,
where `TIME` is an ISO 8601 formatted time, to run the agent in backfill mode.
If you use this option, the agent analyzes all images pushed between that
time and the current time when the agent starts, then exits.

For example:

```console
$ docker run \
  --mount type=bind,src=/var/opt/artifactory-agent,target=/opt/artifactory-agent/data \
  docker/artifactory-agent:v1 --backfill-from=2022-04-10T10:00:00Z
```

When running a backfill multiple times, the agent won't analyze images that
it's already analyzed. To force re-analysis, provide the `--force` command
line flag.

### View analysis results

You can view the image analysis results in the Docker Scout Dashboard.

1. Go to [Images page](https://scout.docker.com/reports/images/) in the Docker Scout Dashboard.

   This page displays the Docker Scout-enabled repositories in your organization.

2. Select the image in the list.
3. Select the tag.

When you have selected a tag, you're taken to the vulnerability report for that
tag. Here, you can select if you want to view all vulnerabilities in the image,
or vulnerabilities introduced in a specific layer. You can also filter
vulnerabilities by severity, and whether or not there's a fix version available.