diff --git a/codedog/chains/pr_summary/base.py b/codedog/chains/pr_summary/base.py
index edef023..e17d0c4 100644
--- a/codedog/chains/pr_summary/base.py
+++ b/codedog/chains/pr_summary/base.py
@@ -50,6 +50,14 @@ class PRSummaryChain(Chain):
 
     @property
     def _chain_type(self) -> str:
+        """
+        Returns the chain type identifier.
+        
+        This method returns a constant string that identifies the chain as a pull request summary chain.
+        
+        Returns:
+            str: The identifier "pull_request_summary_chain".
+        """
         return "pull_request_summary_chain"
 
     @property
diff --git a/codedog/utils/langchain_utils.py b/codedog/utils/langchain_utils.py
index 5954b3c..375d9d7 100644
--- a/codedog/utils/langchain_utils.py
+++ b/codedog/utils/langchain_utils.py
@@ -28,7 +28,15 @@ def load_gpt_llm() -> BaseChatModel:
 
 @lru_cache(maxsize=1)
 def load_gpt4_llm():
-    """Load GPT 4 Model. Make sure your key have access to GPT 4 API. call this function won't check it."""
+    """
+    Load and return a GPT-4 language model instance.
+    
+    This function creates a GPT-4 model based on the environment configuration. If the
+    'AZURE_OPENAI' variable is set, it initializes an AzureChatOpenAI instance configured
+    with Azure-specific parameters; otherwise, it creates a ChatOpenAI instance using the
+    OpenAI API. Note that the function does not verify if the provided API key has access
+    to GPT-4.
+    """
     if env.get("AZURE_OPENAI"):
         llm = AzureChatOpenAI(
             openai_api_type="azure",
diff --git a/tests/conftest.py b/tests/conftest.py
index 1ba0915..28d2377 100644
--- a/tests/conftest.py
+++ b/tests/conftest.py
@@ -3,7 +3,20 @@
 
 @pytest.fixture
 def mock_pull_request():
-    """Create a mock PullRequest object for testing."""
+    """
+    Creates and returns a mock Pull Request object for testing.
+    
+    This fixture creates a MagicMock instance configured with preset attributes:
+        - pull_request_id: 123
+        - repository_id: 456
+        - pull_request_number: 42
+        - title: "Test PR"
+        - body: "PR description"
+        - url: "https://github.com/test/repo/pull/42"
+        - repository_name: "test/repo"
+    
+    The object's json method is set to return an empty JSON object string ("{}").
+    """
     mock_pr = MagicMock()
     mock_pr.pull_request_id = 123
     mock_pr.repository_id = 456
@@ -17,7 +30,13 @@ def mock_pull_request():
 
 @pytest.fixture
 def mock_llm():
-    """Create a mock LLM for testing."""
+    """
+    Create a mock language model for unit testing.
+    
+    Returns a MagicMock instance with a stubbed invoke method that always returns a
+    dictionary containing a test response. This fixture simulates a language model
+    for testing purposes.
+    """
     mock = MagicMock()
     mock.invoke.return_value = {"text": "Test response"}
     return mock 
\ No newline at end of file
diff --git a/tests/integration/test_end_to_end.py b/tests/integration/test_end_to_end.py
index ebd33f0..49db3a8 100644
--- a/tests/integration/test_end_to_end.py
+++ b/tests/integration/test_end_to_end.py
@@ -12,6 +12,15 @@ class TestEndToEndFlow(unittest.TestCase):
     @patch('langchain_openai.chat_models.ChatOpenAI')
     def test_github_to_report_flow(self, mock_chat_openai, mock_github):
         # Setup mocks
+        """
+        Test the GitHub pull request flow for generating a report.
+        
+        This test simulates the end-to-end process of handling a GitHub pull request by creating mock
+        repository and pull request objects, configuring mocked language models, and patching the summary
+        and review chain factories. It verifies that the pull request is correctly summarized and reviewed,
+        and that the reporter compiles the report as expected. The test also asserts that the chain
+        factories and their respective chain calls are invoked exactly once.
+        """
         mock_github_client = MagicMock()
         mock_github.return_value = mock_github_client
         
diff --git a/tests/unit/chains/test_pr_summary_chain.py b/tests/unit/chains/test_pr_summary_chain.py
index 78f455d..e9ca379 100644
--- a/tests/unit/chains/test_pr_summary_chain.py
+++ b/tests/unit/chains/test_pr_summary_chain.py
@@ -9,6 +9,13 @@
 class TestPRSummaryChain(unittest.TestCase):
     def setUp(self):
         # Mock LLM
+        """
+        Set up test fixtures for PRSummaryChain tests.
+        
+        Initializes mocks for the language model and chain dependencies, creates a real output
+        parser, and instantiates a PRSummaryChain object. Also configures a mock pull request and
+        patches the processor to return predefined file changes and summary materials for testing.
+        """
         self.mock_llm = MagicMock(spec=BaseLanguageModel)
         
         # Mock chains
@@ -35,6 +42,13 @@ def setUp(self):
         # Create a real parser instead of a MagicMock
         class TestParser(BaseOutputParser):
             def parse(self, text):
+                """
+                Parses input text to produce a default pull request summary.
+                
+                This stub implementation ignores the input text and always returns a PRSummary
+                with preset values for testing. The summary has an overview of "Parser result",
+                a pull request type set to feature, and major files containing "src/main.py".
+                """
                 return PRSummary(
                     overview="Parser result",
                     pr_type=PRType.feature,
@@ -42,6 +56,11 @@ def parse(self, text):
                 )
                 
             def get_format_instructions(self):
+                """
+                Return the format instructions.
+                
+                This method returns a static string outlining the expected format for outputs.
+                """
                 return "Format instructions"
                 
         # Create chain with a real parser
@@ -78,6 +97,11 @@ def test_process_code_summary_inputs(self):
         
     def test_call(self):
         # Mock run manager
+        """
+        Tests the _call method of PRSummaryChain.
+        
+        Mocks a run manager and verifies that the code summary chain and PR summary chain are each invoked exactly once when calling _call with a mock pull request. The test also asserts that the returned dictionary contains 'pr_summary' and 'code_summaries' keys, with the latter holding exactly one code summary.
+        """
         mock_run_manager = MagicMock()
         mock_run_manager.get_child.return_value = MagicMock()
         
@@ -105,9 +129,26 @@ def test_output_parser_failure(self, mock_translate_chain):
         # Create a failing parser
         class FailingParser(BaseOutputParser):
             def parse(self, text):
+                """
+                Parses the input text.
+                
+                This stub implementation always raises a ValueError to indicate that parsing has failed.
+                
+                Args:
+                    text: The input text to be parsed.
+                
+                Raises:
+                    ValueError: Always raised to signal a parsing error.
+                """
                 raise ValueError("Parsing error")
                 
             def get_format_instructions(self):
+                """
+                Return the format instructions.
+                
+                Returns:
+                    str: A static string indicating the format instructions.
+                """
                 return "Format instructions"
         
         # Create a parser instance
diff --git a/tests/unit/processors/test_pull_request_processor.py b/tests/unit/processors/test_pull_request_processor.py
index e39dc4f..778a44b 100644
--- a/tests/unit/processors/test_pull_request_processor.py
+++ b/tests/unit/processors/test_pull_request_processor.py
@@ -64,6 +64,10 @@ def test_get_diff_code_files(self):
         self.assertEqual(files[0].full_name, "src/main.py")
         
     def test_build_change_summaries(self):
+        """Test conversion of inputs and outputs to ChangeSummary objects.
+        
+            Verifies that build_change_summaries returns a single ChangeSummary with the expected file name and summary text.
+            """
         inputs = [
             {"name": "src/main.py", "language": "python", "content": "diff content"}
         ]
diff --git a/tests/unit/retrievers/test_github_retriever.py b/tests/unit/retrievers/test_github_retriever.py
index c70bf7b..0e6c54e 100644
--- a/tests/unit/retrievers/test_github_retriever.py
+++ b/tests/unit/retrievers/test_github_retriever.py
@@ -9,6 +9,14 @@
 class TestGithubRetriever(unittest.TestCase):
     def setUp(self):
         # Mock Github client and related objects
+        """
+        Initializes test fixtures for GithubRetriever.
+        
+        Sets up mock objects for the GitHub client, repository, and pull request, including
+        simulated commit, file, and issue data. These fixtures are used to instantiate a
+        GithubRetriever and support various test scenarios such as verifying retriever type,
+        parsing issue numbers, handling errors, and processing changed files.
+        """
         self.mock_github = MagicMock(spec=Github)
         self.mock_repo = MagicMock(spec=GHRepo)
         self.mock_pr = MagicMock(spec=GHPullRequest)
@@ -146,6 +154,12 @@ def test_error_handling(self):
 
     def test_empty_pr(self):
         # Test PR with no files
+        """
+        Test that retrieving changed files returns an empty list for a pull request with no files.
+        
+        Sets the internal _changed_files list to empty and confirms that the changed_files
+        property reflects this state by returning an empty list.
+        """
         self.retriever._changed_files = []
         
         # Verify files list is empty
diff --git a/tests/unit/utils/test_langchain_utils.py b/tests/unit/utils/test_langchain_utils.py
index bdc64bc..b12e651 100644
--- a/tests/unit/utils/test_langchain_utils.py
+++ b/tests/unit/utils/test_langchain_utils.py
@@ -12,7 +12,12 @@
 @unittest.skipUnless(HAS_OPENAI, "OpenAI not available")
 class TestLangchainUtils(unittest.TestCase):
     def test_module_imports(self):
-        """Simple test to verify imports work"""
+        """
+        Test that the langchain_utils module is importable and exposes required functions.
+        
+        This test imports the langchain_utils module from the codedog.utils package and asserts the
+        presence of both the load_gpt_llm and load_gpt4_llm functions to confirm the expected API.
+        """
         # This is a basic test to check that our module exists and can be imported
         from codedog.utils import langchain_utils
         self.assertTrue(hasattr(langchain_utils, 'load_gpt_llm'))
@@ -20,7 +25,12 @@ def test_module_imports(self):
         
     @patch('codedog.utils.langchain_utils.env')
     def test_load_gpt_llm_functions(self, mock_env):
-        """Test that the load functions access environment variables"""
+        """
+        Verify that importing load_gpt_llm does not trigger any environment variable access.
+        
+        This test imports load_gpt_llm and asserts that no calls are made to env.get, ensuring
+        that environment variables are not accessed as a side effect of the import.
+        """
         from codedog.utils.langchain_utils import load_gpt_llm
         
         # Mock the env.get calls