diff --git a/src/agentscope/parsers/code_block_parser.py b/src/agentscope/parsers/code_block_parser.py
index df6341123..627d89a1f 100644
--- a/src/agentscope/parsers/code_block_parser.py
+++ b/src/agentscope/parsers/code_block_parser.py
@@ -1,5 +1,7 @@
 # -*- coding: utf-8 -*-
 """Model response parser class for Markdown code block."""
+from typing import Optional
+
 from agentscope.models import ModelResponse
 from agentscope.parsers import ParserBase
 
@@ -22,15 +24,38 @@ class MarkdownCodeBlockParser(ParserBase):
     format_instruction: str = (
         "You should generate {language_name} code in a {language_name} fenced "
         "code block as follows: \n```{language_name}\n"
-        "${{your_{language_name}_code}}\n```"
+        "{content_hint}\n```"
     )
     """The instruction for the format of the code block."""
 
-    def __init__(self, language_name: str) -> None:
+    def __init__(
+        self,
+        language_name: str,
+        content_hint: Optional[str] = None,
+    ) -> None:
+        """Initialize the parser with the language name and the optional
+        content hint.
+
+        Args:
+            language_name (`str`):
+               The name of the language, which will be used
+               in ```{language_name}
+           content_hint (`Optional[str]`, defaults to `None`):
+               The hint used to remind LLM what should be fill between the
+               tags. If not provided, the default content hint
+               "${{your_{language_name}_code}}" will be used.
+        """
         self.name = self.name.format(language_name=language_name)
         self.tag_begin = self.tag_begin.format(language_name=language_name)
+
+        if content_hint is None:
+            self.content_hint = f"${{your_{language_name}_code}}"
+        else:
+            self.content_hint = content_hint
+
         self.format_instruction = self.format_instruction.format(
             language_name=language_name,
+            content_hint=self.content_hint,
         ).strip()
 
     def parse(self, response: ModelResponse) -> ModelResponse:
diff --git a/tests/parser_test.py b/tests/parser_test.py
index 0adf3d728..384ef64cf 100644
--- a/tests/parser_test.py
+++ b/tests/parser_test.py
@@ -95,6 +95,13 @@ def setUp(self) -> None:
             "${your_python_code}\n"
             "```"
         )
+        self.instruction_code_with_hint = (
+            "You should generate python code in a python fenced code block as "
+            "follows: \n"
+            "```python\n"
+            "abc\n"
+            "```"
+        )
         self.gt_code = """\nprint("Hello, world!")\n"""
 
     def test_markdownjsondictparser(self) -> None:
@@ -151,6 +158,22 @@ def test_markdowncodeblockparser(self) -> None:
 
         self.assertEqual(res.parsed, self.gt_code)
 
+    def test_markdowncodeblockparser_with_hint(self) -> None:
+        """Test for MarkdownCodeBlockParser"""
+        parser = MarkdownCodeBlockParser(
+            language_name="python",
+            content_hint="abc",
+        )
+
+        self.assertEqual(
+            parser.format_instruction,
+            self.instruction_code_with_hint,
+        )
+
+        res = parser.parse(self.res_code)
+
+        self.assertEqual(res.parsed, self.gt_code)
+
     def test_multitaggedcontentparser(self) -> None:
         """Test for MultiTaggedContentParser"""
         parser = MultiTaggedContentParser(