From 4fe4becdd5a42e8e4661cc30c6e5a74f3b90614f Mon Sep 17 00:00:00 2001
From: ydy0615 <allenyuan410@gmail.com>
Date: Thu, 19 Feb 2026 11:14:43 +0800
Subject: [PATCH] fix(backend): filter out <br> tags from prompt context

Filter out potential web-scraping or legacy artifacts like <br>, <br/>, <br\> from prefix/suffix before using in model completion context
---
 backend/prompt.py | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/backend/prompt.py b/backend/prompt.py
index 6c35132..10ed6db 100644
--- a/backend/prompt.py
+++ b/backend/prompt.py
@@ -37,8 +37,13 @@ def _sanitize_language_id(language_id: str) -> str:
 def _prepare_context(prefix: str, suffix: str) -> Tuple[str, str]:
     """
     Prepare prefix/suffix for model completion context.
+    Filter out potential web-scraping or legacy artifacts like <br>, <br/>, <br\>.
     """
-    return prefix, suffix
+    import re
+    br_pattern = re.compile(r'<br\s*/?\s*\\?>', re.IGNORECASE)
+    clean_prefix = br_pattern.sub('', prefix or "")
+    clean_suffix = br_pattern.sub('', suffix or "")
+    return clean_prefix, clean_suffix
 
 
 def prepare_prompt_context(prefix: str, suffix: str) -> Tuple[str, str]: