diff --git a/config/locales/client.en.yml b/config/locales/client.en.yml
index d5b555fc..7b6d7f35 100644
--- a/config/locales/client.en.yml
+++ b/config/locales/client.en.yml
@@ -99,9 +99,12 @@ en:
             hide_topic:
               label: "Hide topic"
               description: "Make topic non visible to the public if triggered"
+            flag_type:
+              label: "Flag type"
+              description: "Type of flag to apply to the post (spam or simply raise for review)"
             flag_post:
-              label: "Send to review"
-              description: "Puts the post into the review queue if triggered, for moderators to triage"
+              label: "Flag post"
+              description: "Flags post (either as spam or for review)"
             model:
               label: "Model"
               description: "Language model used for triage"
diff --git a/config/locales/server.en.yml b/config/locales/server.en.yml
index 76a83e33..c42deb03 100644
--- a/config/locales/server.en.yml
+++ b/config/locales/server.en.yml
@@ -1,5 +1,9 @@
 en:
   discourse_automation:
+    ai:
+      flag_types:
+        review: "Add post to review queue"
+        spam: "Flag as spam and hide post"
     scriptables:
       llm_triage:
         title: Triage posts using AI
diff --git a/discourse_automation/llm_triage.rb b/discourse_automation/llm_triage.rb
index b8b52fb4..ad90d361 100644
--- a/discourse_automation/llm_triage.rb
+++ b/discourse_automation/llm_triage.rb
@@ -21,6 +21,13 @@ if defined?(DiscourseAutomation)
     field :tags, component: :tags
     field :hide_topic, component: :boolean
     field :flag_post, component: :boolean
+    field :flag_type,
+          component: :choices,
+          required: false,
+          extra: {
+            content: DiscourseAi::Automation.flag_types,
+          },
+          default: "review"
     field :canned_reply, component: :message
     field :canned_reply_user, component: :user
 
@@ -41,6 +48,7 @@ if defined?(DiscourseAutomation)
       tags = fields.dig("tags", "value")
       hide_topic = fields.dig("hide_topic", "value")
       flag_post = fields.dig("flag_post", "value")
+      flag_type = fields.dig("flag_type", "value")
 
       begin
         RateLimiter.new(
@@ -68,6 +76,7 @@ if defined?(DiscourseAutomation)
           canned_reply_user: canned_reply_user,
           hide_topic: hide_topic,
           flag_post: flag_post,
+          flag_type: flag_type.to_s.to_sym,
           automation: self.automation,
         )
       rescue => e
diff --git a/lib/automation.rb b/lib/automation.rb
index 590bbbe2..e6ed22a2 100644
--- a/lib/automation.rb
+++ b/lib/automation.rb
@@ -2,6 +2,12 @@
 
 module DiscourseAi
   module Automation
+    def self.flag_types
+      [
+        { id: "review", translated_name: I18n.t("discourse_automation.ai.flag_types.review") },
+        { id: "spam", translated_name: I18n.t("discourse_automation.ai.flag_types.spam") },
+      ]
+    end
     def self.available_models
       values = DB.query_hash(<<~SQL)
         SELECT display_name AS translated_name, id AS id
diff --git a/lib/automation/llm_triage.rb b/lib/automation/llm_triage.rb
index 9e7c57b9..103119e7 100644
--- a/lib/automation/llm_triage.rb
+++ b/lib/automation/llm_triage.rb
@@ -14,6 +14,7 @@ module DiscourseAi
         canned_reply_user: nil,
         hide_topic: nil,
         flag_post: nil,
+        flag_type: nil,
         automation: nil
       )
         if category_id.blank? && tags.blank? && canned_reply.blank? && hide_topic.blank? &&
@@ -65,9 +66,6 @@ module DiscourseAi
           post.topic.update!(visible: false) if hide_topic
 
           if flag_post
-            reviewable =
-              ReviewablePost.needs_review!(target: post, created_by: Discourse.system_user)
-
             score_reason =
               I18n
                 .t("discourse_automation.scriptables.llm_triage.flagged_post")
@@ -75,12 +73,25 @@ module DiscourseAi
                 .sub("%%AUTOMATION_ID%%", automation&.id.to_s)
                 .sub("%%AUTOMATION_NAME%%", automation&.name.to_s)
 
-            reviewable.add_score(
-              Discourse.system_user,
-              ReviewableScore.types[:needs_approval],
-              reason: score_reason,
-              force_review: true,
-            )
+            if flag_type == :spam
+              PostActionCreator.new(
+                Discourse.system_user,
+                post,
+                PostActionType.types[:spam],
+                message: score_reason,
+                queue_for_review: true,
+              ).perform
+            else
+              reviewable =
+                ReviewablePost.needs_review!(target: post, created_by: Discourse.system_user)
+
+              reviewable.add_score(
+                Discourse.system_user,
+                ReviewableScore.types[:needs_approval],
+                reason: score_reason,
+                force_review: true,
+              )
+            end
           end
         end
       end
diff --git a/spec/lib/modules/automation/llm_triage_spec.rb b/spec/lib/modules/automation/llm_triage_spec.rb
index 36bc5f32..5fa76309 100644
--- a/spec/lib/modules/automation/llm_triage_spec.rb
+++ b/spec/lib/modules/automation/llm_triage_spec.rb
@@ -1,6 +1,7 @@
 # frozen_string_literal: true
 describe DiscourseAi::Automation::LlmTriage do
   fab!(:post)
+  fab!(:reply) { Fabricate(:post, topic: post.topic, user: Fabricate(:user)) }
   fab!(:llm_model)
 
   def triage(**args)
@@ -92,6 +93,23 @@ describe DiscourseAi::Automation::LlmTriage do
     expect(reviewable.reviewable_scores.first.reason).to include("bad")
   end
 
+  it "can handle spam flags" do
+    DiscourseAi::Completions::Llm.with_prepared_responses(["bad"]) do
+      triage(
+        post: post,
+        model: "custom:#{llm_model.id}",
+        system_prompt: "test %%POST%%",
+        search_for_text: "bad",
+        flag_post: true,
+        flag_type: :spam,
+        automation: nil,
+      )
+    end
+
+    expect(post.reload).to be_hidden
+    expect(post.topic.reload.visible).to eq(false)
+  end
+
   it "can handle garbled output from LLM" do
     DiscourseAi::Completions::Llm.with_prepared_responses(["Bad.\n\nYo"]) do
       triage(