画像生成時のレスポンスをユーザーの言語に合わせてローカライズ (#14)

- エージェントの指示にユーザーの言語で応答するルールを追加
- 画像生成プロンプトもユーザーの言語で記述するよう指示を追加
- generate_image.pyのハードコードされた英語メッセージを除去し、
  エージェントがユーザーの言語で応答できるようnoteフィールドに変更

https://claude.ai/code/session_01QpjLp4XacuMRUZ9GT4RSZn

Co-authored-by: Claude <noreply@anthropic.com>

Author: danishi

app/main.py

@@ -174,13 +174,18 @@ async def _populate_session_from_thread(
 - If asked to summarize a thread, list each person's key points by name.
 - Do NOT include the `[Speaker: ...]` tag in your replies.
 
+### Language
+- Always reply in the same language the user is using.
+- If the user writes in Japanese, reply in Japanese. If in English, reply in English. And so on for any language.
+
 ### Image Generation
 - When the user asks you to create, draw, generate, or design an image, use the `generate_image` tool.
 - Available models:
   - `gemini-3.1-flash-image-preview` (Nanobanana 2): Fast generation (default)
   - `gemini-3-pro-image-preview` (Nanobanana Pro): Higher quality
 - If the user requests a specific model or quality level, set the `model` parameter accordingly.
 - Write a detailed, descriptive prompt for best results.
+- IMPORTANT: Write the image generation prompt in the same language the user used. For example, if the user asks in Japanese, write the prompt in Japanese. This ensures the generated image contains text and cultural elements appropriate to the user's language.
 - Generated images will be automatically uploaded to the Slack thread.
 
 ### Formatting Rules

app/tools/generate_image.py

@@ -80,7 +80,8 @@ def call_gemini():
     if not images:
         return {
             "status": "no_image_generated",
-            "text": "\n".join(text_parts) if text_parts else "No image was generated.",
+            "text": "\n".join(text_parts) if text_parts else "",
+            "note": "No image was produced. Tell the user in their language.",
         }
 
     # Store images for the main handler to upload to Slack
@@ -92,5 +93,6 @@ def call_gemini():
         "status": "success",
         "model": image_model,
         "image_count": len(images),
-        "text": "\n".join(text_parts) if text_parts else f"{len(images)} image(s) generated successfully.",
+        "text": "\n".join(text_parts) if text_parts else "",
+        "note": f"{len(images)} image(s) generated. Inform the user in their language.",
     }