diff --git a/core/src/main/resources/hudson/tasks/test/AggregatedTestResultPublisher/config.jelly b/core/src/main/resources/hudson/tasks/test/AggregatedTestResultPublisher/config.jelly
index 6c19ab89819766473d27c3380ab86edbe083b170..8384829cc51965ae100fd11a17ffe21f93b6253f 100644
--- a/core/src/main/resources/hudson/tasks/test/AggregatedTestResultPublisher/config.jelly
+++ b/core/src/main/resources/hudson/tasks/test/AggregatedTestResultPublisher/config.jelly
@@ -1,6 +1,14 @@
 <j:jelly xmlns:j="jelly:core" xmlns:st="jelly:stapler" xmlns:d="jelly:define" xmlns:l="/lib/layout" xmlns:t="/lib/hudson" xmlns:f="/lib/form">
-  <f:entry title="Jobs to aggregate">
-    <f:textbox name="aggragatedTestResult.jobs" value="${instance.jobs}"
-               checkUrl="'${rootURL}/publisher/AggregatedTestResultPublisher/check?value='+escape(this.value)"/>
-  </f:entry>
+  <f:nested>
+    <table style="width:100%">
+      <f:optionalBlock name="aggragatedTestResult.specify" negative="true"
+              title="${%Automatically aggregate all downstream tests}"
+              help="/help/tasks/aggregate-test/auto-aggregate.html">
+        <f:entry title="Jobs to aggregate">
+          <f:textbox name="aggragatedTestResult.jobs" value="${instance.jobs}"
+                     checkUrl="'${rootURL}/publisher/AggregatedTestResultPublisher/check?value='+escape(this.value)"/>
+        </f:entry>        
+      </f:optionalBlock>
+    </table>
+  </f:nested>
 </j:jelly>
\ No newline at end of file
diff --git a/war/resources/help/tasks/aggregate-test/auto-aggregate.html b/war/resources/help/tasks/aggregate-test/auto-aggregate.html
new file mode 100644
index 0000000000000000000000000000000000000000..600e046515e1bcaa4e4f43422ea9866f875db7eb
--- /dev/null
+++ b/war/resources/help/tasks/aggregate-test/auto-aggregate.html
@@ -0,0 +1,11 @@
+<div>
+    When checked, Hudson simply aggregates tests from
+    all the transitive downstream jobs of this project. This is convenient
+    as you don't have to individually list them and keep them up to date
+    as you add/remove jobs. This normally works just fine, as your test
+    jobs are normally set up as downstream jobs of the build job.
+
+    <p>
+    If for some reasons the auto aggregation doesn't work, you can always
+    manually specify a list of jobs to be aggregated.
+</div>
\ No newline at end of file
diff --git a/war/resources/help/tasks/aggregate-test/manual-list.html b/war/resources/help/tasks/aggregate-test/manual-list.html
new file mode 100644
index 0000000000000000000000000000000000000000..20378b5acf9590fb68ee9912453c802d6455ffdd
--- /dev/null
+++ b/war/resources/help/tasks/aggregate-test/manual-list.html
@@ -0,0 +1,4 @@
+<div>
+    Specify a list of job names manually here. Multiple names can be separated by
+    camma, like "<tt>foo, bar</tt>"
+</div>
\ No newline at end of file