diff --git a/.opencode/command/commit.md b/.opencode/command/commit.md
index c318ed54b1a..8e9346ebc88 100644
--- a/.opencode/command/commit.md
+++ b/.opencode/command/commit.md
@@ -1,6 +1,7 @@
 ---
 description: git commit and push
 model: opencode/glm-4.6
+subtask: true
 ---
 
 commit and push
diff --git a/packages/opencode/src/session/prompt.ts b/packages/opencode/src/session/prompt.ts
index 3be4c45fd60..4ae7469a309 100644
--- a/packages/opencode/src/session/prompt.ts
+++ b/packages/opencode/src/session/prompt.ts
@@ -396,6 +396,30 @@ export namespace SessionPrompt {
             },
           } satisfies MessageV2.ToolPart)
         }
+
+        // Add synthetic user message to prevent certain reasoning models from erroring
+        // If we create assistant messages w/ out user ones following mid loop thinking signatures
+        // will be missing and it can cause errors for models like gemini for example
+        const summaryUserMsg: MessageV2.User = {
+          id: Identifier.ascending("message"),
+          sessionID,
+          role: "user",
+          time: {
+            created: Date.now(),
+          },
+          agent: lastUser.agent,
+          model: lastUser.model,
+        }
+        await Session.updateMessage(summaryUserMsg)
+        await Session.updatePart({
+          id: Identifier.ascending("part"),
+          messageID: summaryUserMsg.id,
+          sessionID,
+          type: "text",
+          text: "Summarize the task tool output above and continue with your task.",
+          synthetic: true,
+        } satisfies MessageV2.TextPart)
+
         continue
       }