feat: add priority input for within-org run ordering

step.sh

@@ -172,6 +172,7 @@ echo "maestro_version: $maestro_version"
 echo "metadata: $metadata"
 echo "name: $name"
 echo "orientation: $orientation"
+echo "priority: $priority"
 echo "report: $report"
 echo "retry: $retry"
 echo "runner_type: $runner_type"
@@ -222,6 +223,7 @@ ${maestro_version:+--maestro-version \"$maestro_version\"} \
 ${name:+--name \"$name\"} \
 ${check_name:+-m \"gh_check_name=$check_name\"} \
 ${orientation:+--orientation \"$orientation\"} \
+${priority:+--priority \"$priority\"} \
 ${report:+--report \"$report\"} \
 ${retry:+--retry \"$retry\"} \
 ${runner_type:+--runner-type \"$runner_type\"} \
@@ -277,6 +279,7 @@ ${maestro_version:+--maestro-version "$maestro_version"} \
 ${name:+--name "$name"} \
 ${check_name:+-m "gh_check_name=$check_name"} \
 ${orientation:+--orientation "$orientation"} \
+${priority:+--priority "$priority"} \
 ${report:+--report "$report"} \
 ${retry:+--retry "$retry"} \
 ${runner_type:+--runner-type "$runner_type"} \

step.yml

@@ -320,6 +320,22 @@ inputs:
       is_expand: true
       is_required: false
 
+  - priority: ""
+    opts:
+      title: "Priority"
+      summary: "Run priority within your organization (low, normal or high). Reorders only your own queued runs; never moves them ahead of other customers. Leave empty for normal."
+      description: |-
+        Decides which of your organization's own queued runs starts first, across
+        your whole concurrency pool, for every flow in this run. It never makes a
+        run start earlier relative to other customers: your organization keeps the
+        same place in the shared queue, and priority only picks which of your runs
+        uses it. Runs queued longer than 6 hours are promoted first, so `low` runs
+        are never starved, and retries keep the run's priority.
+
+        Options: `low`, `normal`, `high`. Leave empty for the default, `normal`.
+      is_expand: true
+      is_required: false
+
   - render_engine: ""
     opts:
       title: "Render Engine"

test/test.bats

@@ -86,7 +86,7 @@ STUB
   # runner may have exported so each test specifies exactly what it wants.
   unset api_key app_file workspace android_device android_api_level ios_device \
         name check_name async google_play debug disable_animations use_beta \
-        env_list metadata download_artifacts json_file cancel_previous \
+        env_list metadata download_artifacts json_file cancel_previous priority \
         include_github_context DCD_STEP_VERSION \
         STUB_STATUS STUB_CLOUD_EXIT STUB_STATUS_FIXTURE STUB_STATUS_RAW
   # ...and the Bitrise env vars the GitHub context is derived from, in case
@@ -214,6 +214,21 @@ teardown() {
   [[ "$output" != *"--cancel-previous"* ]]
 }
 
+@test "passes --priority only when priority is set" {
+  export api_key="k"
+
+  export priority="high"
+  run bash "${TEST_DIR}/step.sh"
+  [ "$status" -eq 0 ]
+  [[ "$output" == *"STUB_CLOUD_CALLED_WITH:"*"--priority high"* ]]
+  [[ "$output" == *'--priority "high"'* ]]
+
+  export priority=""
+  run bash "${TEST_DIR}/step.sh"
+  [ "$status" -eq 0 ]
+  [[ "$output" != *"--priority"* ]]
+}
+
 @test "boolean flags appear only when their inputs are true" {
   export api_key="k"
   export google_play="true"