Capture Pantheon products

Screens for each area that Pantheon’s docs show: the dashboard, Content Publisher, P1, and Pantheon’s WordPress plugins and Drupal modules. For how many images each area has and where its releases come from, see Pantheon docs coverage.

Area Preset Starter brief Status
Pantheon WordPress plugins wordpress-admin briefs/pantheon-wordpress-plugins.json Verified: 5 screens on a local site
Pantheon Drupal modules drupal-admin briefs/pantheon-drupal-modules.json Verified: 7 screens on a local site
Content Publisher and P1 dashboard (content.pantheon.io) content-publisher briefs/content-publisher.json Verified: 4 screens
P1 editor p1-editor briefs/p1-editor.json Verified
Pantheon dashboard (dashboard.pantheon.io) pantheon-dashboard briefs/pantheon-dashboard.json Verified: 16 screens
Google Docs add-on gdocs-addon briefs/gdocs-addon.json, briefs/gdocs-addon-connected.json Verified: 7 screens
Word add-in word-addin briefs/word-addin.json, briefs/word-addin-connected.json Verified: 4 screens
Terminal output none none Show it as text, not an image

Pantheon WordPress plugins

  1. Install and activate the plugins on a local or Dev site, for example the WordPress Studio site from Capture WordPress, Drupal, and other apps:

    wp plugin install pantheon-content-publisher wp-saml-auth wp-native-php-sessions pantheon-hud pantheon-advanced-page-cache solr-power --activate
wp plugin install https://github.com/pantheon-systems/wp-tls-compatibility-checker/archive/refs/tags/1.0.0.zip --activate

  2. Capture with the wordpress-admin config and briefs/pantheon-wordpress-plugins.json:

      content-publisher                             OK
  wp-saml-auth                                  OK
  native-php-sessions                           OK
  tls-compatibility-checker                     OK
  pantheon-hud                                  OK

Done: 5/5 captured, 0 failed, 0 with HTTP >= 400
    The Pantheon Content Publisher plugin's page in WordPress admin: Get started with Content Publisher by connecting a collection, with Connect existing collection and Create new collection.
    content-publisher: the plugin's start page, before a collection is connected.
Plugin Screen Route Where it renders
Pantheon Content Publisher Content Publisher /wp-admin/admin.php?page=pantheon-content-publisher Any site
WP SAML Auth Settings > WP SAML Auth /wp-admin/options-general.php?page=wp-saml-auth-settings Any site
WordPress Native PHP Sessions Tools > Pantheon Sessions /wp-admin/tools.php?page=pantheon-sessions Any site
WP TLS Compatibility Checker Tools > TLS Compatibility Checker /wp-admin/tools.php?page=tls-compatibility-checker Any site
Pantheon HUD Admin bar menu every admin page (#wp-admin-bar-pantheon-hud) Any site; shows only “local” off Pantheon
Pantheon MU plugin Settings > Pantheon Page Cache /wp-admin/options-general.php?page=pantheon-cache Pantheon only
Pantheon Advanced Page Cache Clear cache admin bar item, max-age panel admin bar, and the Pantheon Page Cache page Pantheon only
Solr Power Solr Power /wp-admin/admin.php?page=solr-power Pantheon only (needs a Solr index)
WP Redis none (WP-CLI only)    

“Pantheon only” screens refused access or were missing on a local site. Capture them from a Pantheon Dev or Multidev environment: set "signIn": "chrome" in the config, open the site’s WordPress admin from the dashboard’s Site Admin button in the dedicated Chrome, and add the routes to a copy of the brief.

Pantheon Drupal modules

  1. Add and enable the modules. With Composer and Drush in the site:

    composer require drush/drush drupal/pantheon_content_publisher drupal/pantheon_secrets drupal/search_api_pantheon
vendor/bin/drush pm:install -y pantheon_content_publisher pantheon_secrets search_api_pantheon

    TLS Checker and Pantheon Domain Masking are on GitHub, not Drupal.org: download their latest release into web/modules/custom/ (as tls_checker and pantheon_domain_masking) and enable them the same way.

  2. Capture with the drupal-admin config and briefs/pantheon-drupal-modules.json (7 of 7 captured on Drupal 11.4).

    Drupal's Pantheon Content Publisher collections page with a Connect a new collection button and an empty collections table.
    content-publisher-collections
Module Screen Route
Pantheon Content Publisher Collections /admin/structure/pantheon-content-publisher-collection
Pantheon Content Publisher Documents /admin/content/pantheon-content-publisher
Pantheon Content Publisher Smart components /admin/structure/pantheon-smart-component
Pantheon Secrets Sync all Pantheon Secrets /admin/config/system/keys/pantheon
Search API Pantheon Search API (the “Pantheon” server type) /admin/config/search/search-api
Pantheon Domain Masking Options /admin/config/pantheon-domain-masking
TLS Checker TLS Compatibility Checker /admin/config/development/tls-checker
Pantheon Advanced Page Cache none  

Content Publisher and the P1 dashboard

Content Publisher’s dashboard and P1’s dashboard are the same app at content.pantheon.io, so one preset covers both. Follow Capture WordPress, Drupal, and other apps: Content Publisher.

The Content Publisher docs show this dashboard with two older sidebars, and the P1 docs show a newer one (Overview, Sites, Content Publisher, Team), so the 32 Content Publisher dashboard images are likely out of date. That’s from comparing the docs’ images with each other; the live dashboard wasn’t compared.

The Pantheon dashboard

The dashboard is in 438 of the 1,003 image references on docs.pantheon.io, more than any other screen.

  1. Use a test workspace. Shots show workspace, site, and team names, so a production workspace would show customer data. Personal settings shots show your own name, email address, and SSH key fingerprints.
  2. Copy examples/pantheon-dashboard.config.example.json. Set params.workspaceId from the dashboard’s URL (dashboard.pantheon.io/workspace/<workspaceId>/home) and, for site shots, params.siteId from a site’s URL (…/cms-site/<siteId>/…). Pick a site that isn’t frozen: a frozen site’s pages redirect to a “frozen” notice. params.env picks the environment (default dev).
  3. Start the dedicated Chrome with node scripts/chrome.mjs --config <file> and sign in through Pantheon’s single sign-on.

  4. Capture with briefs/pantheon-dashboard.json:

      workspace-home                                OK
  sites                                         OK
  team                                          OK
  upstreams                                     OK
  workspace-settings                            OK
  site-code                                     OK
  site-database                                 OK
  site-backups                                  OK
  site-domains                                  OK
  site-status                                   OK

  personal-profile                              OK
  …
  personal-tokens                               OK

Done: 16/16 captured, 0 failed, 0 with HTTP >= 400
Screen Route Check
Workspace home /workspace/<workspaceId>/home global navigation, main h1
Sites /workspace/<workspaceId>/sites main h1 contains “Sites”
Team /workspace/<workspaceId>/team main h1 contains “Workspace Team”
Upstreams /workspace/<workspaceId>/upstreams main h1 contains “Custom Upstreams”
Workspace settings /workspace/<workspaceId>/settings/profile main h1 contains “Workspace settings”
Environment: Code /workspace/<workspaceId>/cms-site/<siteId>/environment/<env>/code site and environment navigation, main h2
Environment: Database & Files …/environment/<env>/database/clone (also import, export, wipe) main h2 contains “Clone database and files”
Environment: Backups …/environment/<env>/backups/history (also schedule) main h2 contains “Backup History”
Environment: Domains & HTTPS …/environment/<env>/domains main h2 contains “Domains & HTTPS”
Environment: Status …/environment/<env>/status main h2 contains “Status”
Personal settings /workspace/<workspaceId>/personal-settings/<page>: profile, login-details, email-notifications, security, ssh-keys, personal-access-token main h1 “User settings” and the page’s main h2

The other tabs follow the same pattern: merge, errors, security, newrelic, and, for Test and Live, deploys. Site-level pages are …/cms-site/<siteId>/settings/details and …/cms-site/<siteId>/workflows. Insights, Autopilot, Edge, Support, and Billing are under /workspace/<workspaceId>/.

The preset hides the maintenance and warning banners, toast notifications, and the support chat. It leaves in the “Upgrade to Next Generation GCDN” callout on site pages, which has no stable selector. The dashboard’s header offers “Try the new dashboard”; these routes and checks are for the current one, and a switch to the new one means checking them again.

The Google Docs add-on

Content Publisher’s Google Docs add-on is in 45 of the 149 images on docs.content.pantheon.io. It’s a Google Workspace add-on: it opens from its button in the side panel on the right of Docs, labeled “Pantheon”, not from the Extensions menu, whose Content Publisher entry only has Help.

  1. Use a test doc: shots show the doc’s content and your Google avatar. For the connect screen, use a doc that isn’t connected to a collection.
  2. Copy examples/gdocs-addon.config.example.json and set params.documentId from the doc’s URL (docs.google.com/document/d/<documentId>/edit).
  3. Start the dedicated Chrome and sign in to Google.
  4. Capture with briefs/gdocs-addon.json.

The openAddon action presses the side panel button (or leaves it alone when the panel is already open), waits for the panel, and checks the panel’s text, for example “Connect this doc to a collection”. It clicks nothing inside the add-on, so nothing gets connected or published, and the panel closes when capture closes its tab. The button is a tab: it reports open with aria-selected, not aria-pressed.

After the doc is connected

briefs/gdocs-addon-connected.json captures six screens of a doc that’s connected to a collection:

Shot Screen Where it renders
addon-connected Page metadata, components, formatting check, and the actions The panel
addon-about-collection About this collection: name, URL, ID, and administrator tools The panel
addon-components Components The panel
addon-formatting-check Formatting check results The panel, about 25 seconds later
addon-preview-publish Preview and publish A window the add-on opens
addon-edit-metadata Edit metadata content.pantheon.io, by its URL
  1. Connect the test doc to a test collection whose site publishes nowhere, for example one whose URL is https://example.com. The add-on’s collection list includes production collections, so check the name, URL, and ID on its confirmation screen before you press Connect to collection.

  2. Capture with briefs/gdocs-addon-connected.json and the same config:

      addon-connected                               OK
  addon-about-collection                        OK
  addon-components                              OK
  addon-formatting-check                        OK
  addon-preview-publish                         OK
  addon-edit-metadata                           OK

Done: 6/6 captured, 0 failed, 0 with HTTP >= 400

addonClick clicks a panel button by its text and waits for the panel’s text to change. addonWindow clicks a button that opens a window, takes the shot of that window, and closes it. Preview and publish only loads from the add-on: opened by its URL, it shows “Failed to load page”. It also shows “Checking content quality issues…” for a few seconds, so the shot waits until that’s gone. No shot clicks Publish, Save, or Disconnect collection, and a test in the suite fails if a brief tells an action to.

The formatting check’s results depend on the doc: the test doc shows “Unsupported Format (289)”.

The add-on draws its buttons as images. Connect to collection has the alt text “Connect to Site”, and other images have none, so a screen reader announces the wrong label or nothing. Capture doesn’t need these buttons, because the doc is connected beforehand, but that’s worth reporting to the Content Publisher team.

The Word add-in

Content Publisher’s Microsoft Word add-in is in 6 of the 149 images on docs.content.pantheon.io. It runs in Word for the web as a task pane, opened from its button on the Home tab.

  1. In the dedicated Chrome, sign in to Microsoft and create a test document with some text in it. On an empty document Word shows a “Draft with Copilot” prompt, which ends up in the shots.
  2. Add the add-in: Home > Add-ins, search for Pantheon Content Publisher, and add it. That accepts Microsoft’s and Pantheon’s terms, so it’s yours to do.
  3. Copy examples/word-addin.config.example.json and set params.documentUrl to the document’s full URL from the address bar.

  4. Capture with briefs/word-addin.json for a document that isn’t connected (the welcome screen), or briefs/word-addin-connected.json for one that is (home, metadata, and settings):

      word-addin-connected                          OK
  word-addin-metadata                           OK
  word-addin-settings                           OK

Done: 3/3 captured, 0 failed, 0 with HTTP >= 400

The task pane is an iframe inside Word’s editor iframe, so the preset’s checks use a list of frames: ["iframe#WacFrame_Word_Inline", "iframe[title=\"Office Add-in Pantheon Content Publisher\"]"]. Word’s own panels, such as Navigation and the Copilot prompt, are inside Word’s frame and aren’t hidden.

Two differences from the Google Docs add-on:

  • The Word add-in signs in to Content Publisher separately. Its collection list can differ from the Google Docs add-on’s: the test account saw only the playground, so the test document is connected to “Content Publisher playground”.
  • Preview and Publish opens the same page as the Google Docs add-on, content.pantheon.io/addon/articles/<id>/preview-publish, so the Google Docs shot (addon-preview-publish) covers it. Word only opens that window from a tab that’s in front; a capture step can do that with { "bringToFront": true }, which takes the screen while the shot runs. For a document connected to the playground, the window’s requests return 403 and it shows “Loading…” without an error. A Word document needs a collection under a connected Microsoft account for the window to load.

Terminal output

34 images on docs.pantheon.io are terminal output, and the Content Publisher docs have 4 terminal screenshots. Use a code block instead: readers can copy it, screen readers can read it, and a release only changes text. The docs for this tool do the same.

Third-party screens

216 images on docs.pantheon.io show other companies’ screens (Google, GitHub, Cloudflare, New Relic, and more). Capture one with the public-site preset if it’s public, or with "signIn": "chrome" and your own sign-in if it isn’t. Their releases don’t come from Pantheon, so the watch doesn’t track them.

This site uses Just the Docs, a documentation theme for Jekyll.