diff --git a/.env.example b/.env.example
index 381729b..fc516a7 100644
--- a/.env.example
+++ b/.env.example
@@ -1,21 +1,28 @@
-# Example .env file for Gemini MCP Server
-# Copy this to .env and update with your actual values
+# Gemini MCP Server Environment Configuration
+# Copy this file to .env and fill in your values
 
-# Your Gemini API key (required)
-# Get one from: https://makersuite.google.com/app/apikey
-GEMINI_API_KEY=your-gemini-api-key-here
+# Required: Google Gemini API Key
+# Get your API key from: https://makersuite.google.com/app/apikey
+GEMINI_API_KEY=your_gemini_api_key_here
 
-# Docker-specific environment variables (optional)
-# These are set automatically by the Docker setup scripts
-# You typically don't need to set these manually
+# Optional: Redis connection URL for conversation memory
+# Defaults to redis://localhost:6379/0
+# For Docker: redis://redis:6379/0
+REDIS_URL=redis://localhost:6379/0
 
-# WORKSPACE_ROOT: Used for Docker path translation
-# Automatically set when using Docker wrapper scripts
-# Example: /Users/username/my-project (macOS/Linux)
-# Example: C:\Users\username\my-project (Windows)
-# WORKSPACE_ROOT=/path/to/your/project
+# Optional: Workspace root directory for file access
+# This should be the HOST path that contains all files Claude might reference
+# Defaults to $HOME for direct usage, auto-configured for Docker
+WORKSPACE_ROOT=/Users/your-username
 
-# MCP_PROJECT_ROOT: Restricts file access to a specific directory
-# If not set, defaults to user's home directory
-# Set this to limit file access to a specific project folder
-# MCP_PROJECT_ROOT=/path/to/allowed/directory
\ No newline at end of file
+# Optional: Logging level (DEBUG, INFO, WARNING, ERROR)
+# DEBUG: Shows detailed operational messages for troubleshooting
+# INFO: Shows general operational messages (default)
+# WARNING: Shows only warnings and errors
+# ERROR: Shows only errors
+LOG_LEVEL=INFO
+
+# Optional: Project root override for file sandboxing
+# If set, overrides the default sandbox directory
+# Use with caution - this controls which files the server can access
+# MCP_PROJECT_ROOT=/path/to/specific/project
\ No newline at end of file
diff --git a/README.md b/README.md
index a556ad1..a9b4f93 100644
--- a/README.md
+++ b/README.md
@@ -186,6 +186,7 @@ This server enables **true AI collaboration** between Claude and Gemini, where t
 - **Claude can respond** with additional information, files, or refined instructions
 - **Claude can work independently** between exchanges - implementing solutions, gathering data, or performing analysis
 - **Claude can return to Gemini** with progress updates and new context for further collaboration
+- **Cross-tool continuation** - Start with one tool (e.g., `analyze`) and continue with another (e.g., `codereview`) using the same conversation thread
 - **Both AIs coordinate their approaches** - questioning assumptions, validating solutions, and building on each other's insights
 - Each conversation maintains full context while only sending incremental updates
 - Conversations are automatically managed with Redis for persistence
@@ -208,12 +209,27 @@ This server enables **true AI collaboration** between Claude and Gemini, where t
 - **Coordinated problem-solving**: Each AI contributes their strengths to complex problems
 - **Context building**: Claude gathers information while Gemini provides deep analysis
 - **Approach validation**: AIs can verify and improve each other's solutions
+- **Cross-tool continuation**: Seamlessly continue conversations across different tools while preserving all context
 - **Asynchronous workflow**: Conversations don't need to be sequential - Claude can work on tasks between exchanges, then return to Gemini with additional context and progress updates
 - **Incremental updates**: Share only new information in each exchange while maintaining full conversation history
 - **Automatic 25K limit bypass**: Each exchange sends only incremental context, allowing unlimited total conversation size
 - Up to 5 exchanges per conversation with 1-hour expiry
 - Thread-safe with Redis persistence across all tools
 
+**Cross-tool continuation example:**
+```
+1. Claude: "Use gemini to analyze /src/auth.py for security issues"
+   → Gemini analyzes and finds vulnerabilities, provides continuation_id
+
+2. Claude: "Use gemini to review the authentication logic thoroughly"
+   → Uses same continuation_id, Gemini sees previous analysis and files
+   → Provides detailed code review building on previous findings  
+
+3. Claude: "Use gemini to help debug the auth test failures"
+   → Same continuation_id, full context from analysis + review
+   → Gemini provides targeted debugging with complete understanding
+```
+
 ## Available Tools
 
 **Quick Tool Selection Guide:**
@@ -837,6 +853,28 @@ Different tools use optimized temperature settings:
 - **`TEMPERATURE_BALANCED`**: `0.5` - Used for general chat (balanced creativity/accuracy)
 - **`TEMPERATURE_CREATIVE`**: `0.7` - Used for deep thinking and architecture (more creative)
 
+### Logging Configuration
+Control logging verbosity via the `LOG_LEVEL` environment variable:
+- **`DEBUG`**: Shows detailed operational messages, tool execution flow, conversation threading
+- **`INFO`**: Shows general operational messages (default)
+- **`WARNING`**: Shows only warnings and errors
+- **`ERROR`**: Shows only errors
+
+**Set in your .env file:**
+```bash
+LOG_LEVEL=DEBUG  # For troubleshooting
+LOG_LEVEL=INFO   # For normal operation (default)
+```
+
+**For Docker:**
+```bash
+# In .env file
+LOG_LEVEL=DEBUG
+
+# Or set directly when starting
+LOG_LEVEL=DEBUG docker compose up
+```
+
 
 ## File Path Requirements
 
diff --git a/docker-compose.yml b/docker-compose.yml
index 38c80af..a24ee78 100644
--- a/docker-compose.yml
+++ b/docker-compose.yml
@@ -34,11 +34,13 @@ services:
       # Use HOME not PWD: Claude needs access to any absolute file path, not just current project,
       # and Claude Code could be running from multiple locations at the same time
       - WORKSPACE_ROOT=${WORKSPACE_ROOT:-${HOME}}
+      - LOG_LEVEL=${LOG_LEVEL:-INFO}
     volumes:
       - ${HOME:-/tmp}:/workspace:ro
     stdin_open: true
     tty: true
-    command: ["sh", "-c", "while true; do sleep 86400; done"]
+    entrypoint: ["python"]
+    command: ["log_monitor.py"]
 
 volumes:
   redis_data:
\ No newline at end of file
diff --git a/server.py b/server.py
index bacea30..eebdc18 100644
--- a/server.py
+++ b/server.py
@@ -47,8 +47,22 @@ from tools import (
 )
 
 # Configure logging for server operations
-# Set to DEBUG level to capture detailed operational messages for troubleshooting
-logging.basicConfig(level=logging.DEBUG, format="%(asctime)s - %(name)s - %(levelname)s - %(message)s")
+# Can be controlled via LOG_LEVEL environment variable (DEBUG, INFO, WARNING, ERROR)
+log_level = os.getenv("LOG_LEVEL", "INFO").upper()
+
+# Configure both console and file logging
+log_format = "%(asctime)s - %(name)s - %(levelname)s - %(message)s"
+logging.basicConfig(level=getattr(logging, log_level, logging.INFO), format=log_format)
+
+# Add file handler for Docker log monitoring
+try:
+    file_handler = logging.FileHandler("/tmp/mcp_server.log")
+    file_handler.setLevel(getattr(logging, log_level, logging.INFO))
+    file_handler.setFormatter(logging.Formatter(log_format))
+    logging.getLogger().addHandler(file_handler)
+except Exception as e:
+    print(f"Warning: Could not set up file logging: {e}")
+
 logger = logging.getLogger(__name__)
 
 # Create the MCP server instance with a unique name identifier
diff --git a/setup-docker.sh b/setup-docker.sh
index 5d7e654..ce7aa3c 100755
--- a/setup-docker.sh
+++ b/setup-docker.sh
@@ -43,6 +43,14 @@ REDIS_URL=redis://redis:6379/0
 # not just files within the current project directory. Additionally, Claude Code
 # could be running from multiple locations at the same time.
 WORKSPACE_ROOT=$HOME
+
+# Logging level (DEBUG, INFO, WARNING, ERROR)
+# DEBUG: Shows detailed operational messages, conversation threading, tool execution flow
+# INFO: Shows general operational messages (default)
+# WARNING: Shows only warnings and errors
+# ERROR: Shows only errors
+# Uncomment and change to DEBUG if you need detailed troubleshooting information
+LOG_LEVEL=INFO
 EOF
     echo "✅ Created .env file with Redis configuration"
     echo ""
@@ -168,6 +176,17 @@ echo "  }"
 echo "}"
 echo "==========================================="
 echo ""
+echo "===== CLAUDE CODE CLI CONFIGURATION ====="
+echo "# Add the MCP server via Claude Code CLI:"
+echo "claude mcp add gemini -s user -- docker exec -i gemini-mcp-server python server.py"
+echo ""
+echo "# List your MCP servers to verify:"
+echo "claude mcp list"
+echo ""
+echo "# Remove if needed:"
+echo "claude mcp remove gemini"
+echo "==========================================="
+echo ""
 
 echo "📁 Config file locations:"
 echo "  macOS: ~/Library/Application Support/Claude/claude_desktop_config.json"
diff --git a/setup.py b/setup.py
index 37b6745..fac6678 100644
--- a/setup.py
+++ b/setup.py
@@ -14,7 +14,7 @@ if readme_path.exists():
 
 setup(
     name="gemini-mcp-server",
-    version="3.0.0",
+    version="3.1.0",
     description="Model Context Protocol server for Google Gemini",
     long_description=long_description,
     long_description_content_type="text/markdown",
diff --git a/tests/test_claude_continuation.py b/tests/test_claude_continuation.py
index a88da10..5db406b 100644
--- a/tests/test_claude_continuation.py
+++ b/tests/test_claude_continuation.py
@@ -241,7 +241,7 @@ class TestClaudeContinuationOffers:
     def test_max_turns_reached_no_continuation_offer(self):
         """Test that no continuation is offered when max turns would be exceeded"""
         # Mock MAX_CONVERSATION_TURNS to be 1 for this test
-        with patch("utils.conversation_memory.MAX_CONVERSATION_TURNS", 1):
+        with patch("tools.base.MAX_CONVERSATION_TURNS", 1):
             request = ContinuationRequest(prompt="Test prompt")
 
             # Check continuation opportunity
diff --git a/tests/test_collaboration.py b/tests/test_collaboration.py
index 6e7fe93..8d653c9 100644
--- a/tests/test_collaboration.py
+++ b/tests/test_collaboration.py
@@ -235,9 +235,9 @@ class TestCollaborationWorkflow:
         )
 
         response = json.loads(result[0].text)
-        assert response["status"] == "requires_clarification", (
-            "Should request clarification when asked about dependencies without package files"
-        )
+        assert (
+            response["status"] == "requires_clarification"
+        ), "Should request clarification when asked about dependencies without package files"
 
         clarification = json.loads(response["content"])
         assert "package.json" in str(clarification["files_needed"]), "Should specifically request package.json"
diff --git a/tests/test_conversation_history_bug.py b/tests/test_conversation_history_bug.py
new file mode 100644
index 0000000..49d2737
--- /dev/null
+++ b/tests/test_conversation_history_bug.py
@@ -0,0 +1,251 @@
+"""
+Test suite for conversation history bug fix
+
+This test verifies that the critical bug where conversation history
+(including file context) was not included when using continuation_id
+has been properly fixed.
+
+The bug was that tools with continuation_id would not see previous
+conversation turns, causing issues like Gemini not seeing files that
+Claude had shared in earlier turns.
+"""
+
+import json
+from unittest.mock import Mock, patch
+
+import pytest
+from pydantic import Field
+
+from tools.base import BaseTool, ToolRequest
+from utils.conversation_memory import ConversationTurn, ThreadContext
+
+
+class FileContextRequest(ToolRequest):
+    """Test request with file support"""
+
+    prompt: str = Field(..., description="Test prompt")
+    files: list[str] = Field(default_factory=list, description="Optional files")
+
+
+class FileContextTool(BaseTool):
+    """Test tool for file context verification"""
+
+    def get_name(self) -> str:
+        return "test_file_context"
+
+    def get_description(self) -> str:
+        return "Test tool for file context"
+
+    def get_input_schema(self) -> dict:
+        return {
+            "type": "object",
+            "properties": {
+                "prompt": {"type": "string"},
+                "files": {"type": "array", "items": {"type": "string"}},
+                "continuation_id": {"type": "string", "required": False},
+            },
+        }
+
+    def get_system_prompt(self) -> str:
+        return "Test system prompt for file context"
+
+    def get_request_model(self):
+        return FileContextRequest
+
+    async def prepare_prompt(self, request) -> str:
+        # Simple prompt preparation that would normally read files
+        # For this test, we're focusing on whether conversation history is included
+        files_context = ""
+        if request.files:
+            files_context = f"\nFiles in current request: {', '.join(request.files)}"
+
+        return f"System: {self.get_system_prompt()}\nUser: {request.prompt}{files_context}"
+
+
+class TestConversationHistoryBugFix:
+    """Test that conversation history is properly included with continuation_id"""
+
+    def setup_method(self):
+        self.tool = FileContextTool()
+
+    @patch("tools.base.get_thread")
+    @patch("tools.base.add_turn")
+    async def test_conversation_history_included_with_continuation_id(self, mock_add_turn, mock_get_thread):
+        """Test that conversation history (including file context) is included when using continuation_id"""
+
+        # Create a thread context with previous turns including files
+        thread_context = ThreadContext(
+            thread_id="test-history-id",
+            created_at="2023-01-01T00:00:00Z",
+            last_updated_at="2023-01-01T00:02:00Z",
+            tool_name="analyze",  # Started with analyze tool
+            turns=[
+                ConversationTurn(
+                    role="assistant",
+                    content="I've analyzed the authentication module and found several security issues.",
+                    timestamp="2023-01-01T00:01:00Z",
+                    tool_name="analyze",
+                    files=["/src/auth.py", "/src/security.py"],  # Files from analyze tool
+                ),
+                ConversationTurn(
+                    role="assistant",
+                    content="The code review shows these files have critical vulnerabilities.",
+                    timestamp="2023-01-01T00:02:00Z",
+                    tool_name="codereview",
+                    files=["/src/auth.py", "/tests/test_auth.py"],  # Files from codereview tool
+                ),
+            ],
+            initial_context={"question": "Analyze authentication security"},
+        )
+
+        # Mock get_thread to return our test context
+        mock_get_thread.return_value = thread_context
+        # Mock add_turn to return success
+        mock_add_turn.return_value = True
+
+        # Mock the model to capture what prompt it receives
+        captured_prompt = None
+
+        with patch.object(self.tool, "create_model") as mock_create_model:
+            mock_model = Mock()
+            mock_response = Mock()
+            mock_response.candidates = [
+                Mock(
+                    content=Mock(parts=[Mock(text="Response with conversation context")]),
+                    finish_reason="STOP",
+                )
+            ]
+
+            def capture_prompt(prompt):
+                nonlocal captured_prompt
+                captured_prompt = prompt
+                return mock_response
+
+            mock_model.generate_content.side_effect = capture_prompt
+            mock_create_model.return_value = mock_model
+
+            # Execute tool with continuation_id
+            arguments = {
+                "prompt": "What should we fix first?",
+                "continuation_id": "test-history-id",
+                "files": ["/src/utils.py"],  # New file for this turn
+            }
+            response = await self.tool.execute(arguments)
+
+            # Verify response succeeded
+            response_data = json.loads(response[0].text)
+            assert response_data["status"] == "success"
+
+            # Verify get_thread was called for history reconstruction
+            mock_get_thread.assert_called_with("test-history-id")
+
+            # Verify the prompt includes conversation history
+            assert captured_prompt is not None
+
+            # Check that conversation history is included
+            assert "=== CONVERSATION HISTORY ===" in captured_prompt
+            assert "Turn 1 (Gemini using analyze)" in captured_prompt
+            assert "Turn 2 (Gemini using codereview)" in captured_prompt
+
+            # Check that file context from previous turns is included
+            assert "📁 Files referenced: /src/auth.py, /src/security.py" in captured_prompt
+            assert "📁 Files referenced: /src/auth.py, /tests/test_auth.py" in captured_prompt
+
+            # Check that previous turn content is included
+            assert "I've analyzed the authentication module and found several security issues." in captured_prompt
+            assert "The code review shows these files have critical vulnerabilities." in captured_prompt
+
+            # Check that continuation instruction is included
+            assert "Continue this conversation by building on the previous context." in captured_prompt
+
+            # Check that current request is still included
+            assert "What should we fix first?" in captured_prompt
+            assert "Files in current request: /src/utils.py" in captured_prompt
+
+    @patch("tools.base.get_thread")
+    async def test_no_history_when_thread_not_found(self, mock_get_thread):
+        """Test graceful handling when thread is not found"""
+
+        # Mock get_thread to return None (thread not found)
+        mock_get_thread.return_value = None
+
+        captured_prompt = None
+
+        with patch.object(self.tool, "create_model") as mock_create_model:
+            mock_model = Mock()
+            mock_response = Mock()
+            mock_response.candidates = [
+                Mock(
+                    content=Mock(parts=[Mock(text="Response without history")]),
+                    finish_reason="STOP",
+                )
+            ]
+
+            def capture_prompt(prompt):
+                nonlocal captured_prompt
+                captured_prompt = prompt
+                return mock_response
+
+            mock_model.generate_content.side_effect = capture_prompt
+            mock_create_model.return_value = mock_model
+
+            # Execute tool with continuation_id for non-existent thread
+            arguments = {"prompt": "Test without history", "continuation_id": "non-existent-thread-id"}
+            response = await self.tool.execute(arguments)
+
+            # Should still succeed but without history
+            response_data = json.loads(response[0].text)
+            assert response_data["status"] == "success"
+
+            # Verify get_thread was called for non-existent thread
+            mock_get_thread.assert_called_with("non-existent-thread-id")
+
+            # Verify the prompt does NOT include conversation history
+            assert captured_prompt is not None
+            assert "=== CONVERSATION HISTORY ===" not in captured_prompt
+            assert "Test without history" in captured_prompt
+
+    async def test_no_history_for_new_conversations(self):
+        """Test that new conversations (no continuation_id) don't get history"""
+
+        captured_prompt = None
+
+        with patch.object(self.tool, "create_model") as mock_create_model:
+            mock_model = Mock()
+            mock_response = Mock()
+            mock_response.candidates = [
+                Mock(
+                    content=Mock(parts=[Mock(text="New conversation response")]),
+                    finish_reason="STOP",
+                )
+            ]
+
+            def capture_prompt(prompt):
+                nonlocal captured_prompt
+                captured_prompt = prompt
+                return mock_response
+
+            mock_model.generate_content.side_effect = capture_prompt
+            mock_create_model.return_value = mock_model
+
+            # Execute tool without continuation_id (new conversation)
+            arguments = {"prompt": "Start new conversation", "files": ["/src/new_file.py"]}
+            response = await self.tool.execute(arguments)
+
+            # Should succeed (may offer continuation for new conversations)
+            response_data = json.loads(response[0].text)
+            assert response_data["status"] in ["success", "continuation_available"]
+
+            # Verify the prompt does NOT include conversation history
+            assert captured_prompt is not None
+            assert "=== CONVERSATION HISTORY ===" not in captured_prompt
+            assert "Start new conversation" in captured_prompt
+            assert "Files in current request: /src/new_file.py" in captured_prompt
+
+            # Should include follow-up instructions for new conversation
+            # (This is the existing behavior for new conversations)
+            assert "If you'd like to ask a follow-up question" in captured_prompt
+
+
+if __name__ == "__main__":
+    pytest.main([__file__])
diff --git a/tests/test_cross_tool_continuation.py b/tests/test_cross_tool_continuation.py
new file mode 100644
index 0000000..ef3861a
--- /dev/null
+++ b/tests/test_cross_tool_continuation.py
@@ -0,0 +1,381 @@
+"""
+Test suite for cross-tool continuation functionality
+
+Tests that continuation IDs work properly across different tools,
+allowing multi-turn conversations to span multiple tool types.
+"""
+
+import json
+from unittest.mock import Mock, patch
+
+import pytest
+from pydantic import Field
+
+from tools.base import BaseTool, ToolRequest
+from utils.conversation_memory import ConversationTurn, ThreadContext
+
+
+class AnalysisRequest(ToolRequest):
+    """Test request for analysis tool"""
+
+    code: str = Field(..., description="Code to analyze")
+
+
+class ReviewRequest(ToolRequest):
+    """Test request for review tool"""
+
+    findings: str = Field(..., description="Analysis findings to review")
+    files: list[str] = Field(default_factory=list, description="Optional files to review")
+
+
+class MockAnalysisTool(BaseTool):
+    """Mock analysis tool for cross-tool testing"""
+
+    def get_name(self) -> str:
+        return "test_analysis"
+
+    def get_description(self) -> str:
+        return "Test analysis tool"
+
+    def get_input_schema(self) -> dict:
+        return {
+            "type": "object",
+            "properties": {
+                "code": {"type": "string"},
+                "continuation_id": {"type": "string", "required": False},
+            },
+        }
+
+    def get_system_prompt(self) -> str:
+        return "Analyze the provided code"
+
+    def get_request_model(self):
+        return AnalysisRequest
+
+    async def prepare_prompt(self, request) -> str:
+        return f"System: {self.get_system_prompt()}\nCode: {request.code}"
+
+
+class MockReviewTool(BaseTool):
+    """Mock review tool for cross-tool testing"""
+
+    def get_name(self) -> str:
+        return "test_review"
+
+    def get_description(self) -> str:
+        return "Test review tool"
+
+    def get_input_schema(self) -> dict:
+        return {
+            "type": "object",
+            "properties": {
+                "findings": {"type": "string"},
+                "continuation_id": {"type": "string", "required": False},
+            },
+        }
+
+    def get_system_prompt(self) -> str:
+        return "Review the analysis findings"
+
+    def get_request_model(self):
+        return ReviewRequest
+
+    async def prepare_prompt(self, request) -> str:
+        return f"System: {self.get_system_prompt()}\nFindings: {request.findings}"
+
+
+class TestCrossToolContinuation:
+    """Test cross-tool continuation functionality"""
+
+    def setup_method(self):
+        self.analysis_tool = MockAnalysisTool()
+        self.review_tool = MockReviewTool()
+
+    @patch("utils.conversation_memory.get_redis_client")
+    async def test_continuation_id_works_across_different_tools(self, mock_redis):
+        """Test that a continuation_id from one tool can be used with another tool"""
+        mock_client = Mock()
+        mock_redis.return_value = mock_client
+
+        # Step 1: Analysis tool creates a conversation with follow-up
+        with patch.object(self.analysis_tool, "create_model") as mock_create_model:
+            mock_model = Mock()
+            mock_response = Mock()
+            mock_response.candidates = [
+                Mock(
+                    content=Mock(
+                        parts=[
+                            Mock(
+                                text="""Found potential security issues in authentication logic.
+
+```json
+{
+  "follow_up_question": "Would you like me to review these security findings in detail?",
+  "suggested_params": {"findings": "Authentication bypass vulnerability detected"},
+  "ui_hint": "Security review recommended"
+}
+```"""
+                            )
+                        ]
+                    ),
+                    finish_reason="STOP",
+                )
+            ]
+            mock_model.generate_content.return_value = mock_response
+            mock_create_model.return_value = mock_model
+
+            # Execute analysis tool
+            arguments = {"code": "function authenticate(user) { return true; }"}
+            response = await self.analysis_tool.execute(arguments)
+            response_data = json.loads(response[0].text)
+
+            assert response_data["status"] == "requires_continuation"
+            continuation_id = response_data["follow_up_request"]["continuation_id"]
+
+        # Step 2: Mock the existing thread context for the review tool
+        # The thread was created by analysis_tool but will be continued by review_tool
+        existing_context = ThreadContext(
+            thread_id=continuation_id,
+            created_at="2023-01-01T00:00:00Z",
+            last_updated_at="2023-01-01T00:01:00Z",
+            tool_name="test_analysis",  # Original tool
+            turns=[
+                ConversationTurn(
+                    role="assistant",
+                    content="Found potential security issues in authentication logic.",
+                    timestamp="2023-01-01T00:00:30Z",
+                    tool_name="test_analysis",  # Original tool
+                    follow_up_question="Would you like me to review these security findings in detail?",
+                )
+            ],
+            initial_context={"code": "function authenticate(user) { return true; }"},
+        )
+
+        # Mock the get call to return existing context for add_turn to work
+        def mock_get_side_effect(key):
+            if key.startswith("thread:"):
+                return existing_context.model_dump_json()
+            return None
+
+        mock_client.get.side_effect = mock_get_side_effect
+
+        # Step 3: Review tool uses the same continuation_id
+        with patch.object(self.review_tool, "create_model") as mock_create_model:
+            mock_model = Mock()
+            mock_response = Mock()
+            mock_response.candidates = [
+                Mock(
+                    content=Mock(
+                        parts=[
+                            Mock(
+                                text="Critical security vulnerability confirmed. The authentication function always returns true, bypassing all security checks."
+                            )
+                        ]
+                    ),
+                    finish_reason="STOP",
+                )
+            ]
+            mock_model.generate_content.return_value = mock_response
+            mock_create_model.return_value = mock_model
+
+            # Execute review tool with the continuation_id from analysis tool
+            arguments = {
+                "findings": "Authentication bypass vulnerability detected",
+                "continuation_id": continuation_id,
+            }
+            response = await self.review_tool.execute(arguments)
+            response_data = json.loads(response[0].text)
+
+            # Should successfully continue the conversation
+            assert response_data["status"] == "success"
+            assert "Critical security vulnerability confirmed" in response_data["content"]
+
+        # Step 4: Verify the cross-tool continuation worked
+        # Should have at least 2 setex calls: 1 from analysis tool follow-up, 1 from review tool add_turn
+        setex_calls = mock_client.setex.call_args_list
+        assert len(setex_calls) >= 2  # Analysis tool creates thread + review tool adds turn
+
+        # Get the final thread state from the last setex call
+        final_thread_data = setex_calls[-1][0][2]  # Last setex call's data
+        final_context = json.loads(final_thread_data)
+
+        assert final_context["thread_id"] == continuation_id
+        assert final_context["tool_name"] == "test_analysis"  # Original tool name preserved
+        assert len(final_context["turns"]) == 2  # Original + new turn
+
+        # Verify the new turn has the review tool's name
+        second_turn = final_context["turns"][1]
+        assert second_turn["role"] == "assistant"
+        assert second_turn["tool_name"] == "test_review"  # New tool name
+        assert "Critical security vulnerability confirmed" in second_turn["content"]
+
+    @patch("utils.conversation_memory.get_redis_client")
+    def test_cross_tool_conversation_history_includes_tool_names(self, mock_redis):
+        """Test that conversation history properly shows which tool was used for each turn"""
+        mock_client = Mock()
+        mock_redis.return_value = mock_client
+
+        # Create a thread context with turns from different tools
+        thread_context = ThreadContext(
+            thread_id="12345678-1234-1234-1234-123456789012",
+            created_at="2023-01-01T00:00:00Z",
+            last_updated_at="2023-01-01T00:03:00Z",
+            tool_name="test_analysis",  # Original tool
+            turns=[
+                ConversationTurn(
+                    role="assistant",
+                    content="Analysis complete: Found 3 issues",
+                    timestamp="2023-01-01T00:01:00Z",
+                    tool_name="test_analysis",
+                ),
+                ConversationTurn(
+                    role="assistant",
+                    content="Review complete: 2 critical, 1 minor issue",
+                    timestamp="2023-01-01T00:02:00Z",
+                    tool_name="test_review",
+                ),
+                ConversationTurn(
+                    role="assistant",
+                    content="Deep analysis: Root cause identified",
+                    timestamp="2023-01-01T00:03:00Z",
+                    tool_name="test_thinkdeep",
+                ),
+            ],
+            initial_context={"code": "test code"},
+        )
+
+        # Build conversation history
+        from utils.conversation_memory import build_conversation_history
+
+        history = build_conversation_history(thread_context)
+
+        # Verify tool names are included in the history
+        assert "Turn 1 (Gemini using test_analysis)" in history
+        assert "Turn 2 (Gemini using test_review)" in history
+        assert "Turn 3 (Gemini using test_thinkdeep)" in history
+        assert "Analysis complete: Found 3 issues" in history
+        assert "Review complete: 2 critical, 1 minor issue" in history
+        assert "Deep analysis: Root cause identified" in history
+
+    @patch("utils.conversation_memory.get_redis_client")
+    @patch("utils.conversation_memory.get_thread")
+    async def test_cross_tool_conversation_with_files_context(self, mock_get_thread, mock_redis):
+        """Test that file context is preserved across tool switches"""
+        mock_client = Mock()
+        mock_redis.return_value = mock_client
+
+        # Create existing context with files from analysis tool
+        existing_context = ThreadContext(
+            thread_id="test-thread-id",
+            created_at="2023-01-01T00:00:00Z",
+            last_updated_at="2023-01-01T00:01:00Z",
+            tool_name="test_analysis",
+            turns=[
+                ConversationTurn(
+                    role="assistant",
+                    content="Analysis of auth.py complete",
+                    timestamp="2023-01-01T00:01:00Z",
+                    tool_name="test_analysis",
+                    files=["/src/auth.py", "/src/utils.py"],
+                )
+            ],
+            initial_context={"code": "authentication code", "files": ["/src/auth.py"]},
+        )
+
+        # Mock get_thread to return the existing context
+        mock_get_thread.return_value = existing_context
+
+        # Mock review tool response
+        with patch.object(self.review_tool, "create_model") as mock_create_model:
+            mock_model = Mock()
+            mock_response = Mock()
+            mock_response.candidates = [
+                Mock(
+                    content=Mock(parts=[Mock(text="Security review of auth.py shows vulnerabilities")]),
+                    finish_reason="STOP",
+                )
+            ]
+            mock_model.generate_content.return_value = mock_response
+            mock_create_model.return_value = mock_model
+
+            # Execute review tool with additional files
+            arguments = {
+                "findings": "Auth vulnerabilities found",
+                "continuation_id": "test-thread-id",
+                "files": ["/src/security.py"],  # Additional file for review
+            }
+            response = await self.review_tool.execute(arguments)
+            response_data = json.loads(response[0].text)
+
+            assert response_data["status"] == "success"
+
+        # Verify files from both tools are tracked in Redis calls
+        setex_calls = mock_client.setex.call_args_list
+        assert len(setex_calls) >= 1  # At least the add_turn call from review tool
+
+        # Get the final thread state
+        final_thread_data = setex_calls[-1][0][2]
+        final_context = json.loads(final_thread_data)
+
+        # Check that the new turn includes the review tool's files
+        review_turn = final_context["turns"][1]  # Second turn (review tool)
+        assert review_turn["tool_name"] == "test_review"
+        assert review_turn["files"] == ["/src/security.py"]
+
+        # Original turn's files should still be there
+        analysis_turn = final_context["turns"][0]  # First turn (analysis tool)
+        assert analysis_turn["files"] == ["/src/auth.py", "/src/utils.py"]
+
+    @patch("utils.conversation_memory.get_redis_client")
+    @patch("utils.conversation_memory.get_thread")
+    def test_thread_preserves_original_tool_name(self, mock_get_thread, mock_redis):
+        """Test that the thread's original tool_name is preserved even when other tools contribute"""
+        mock_client = Mock()
+        mock_redis.return_value = mock_client
+
+        # Create existing thread from analysis tool
+        existing_context = ThreadContext(
+            thread_id="test-thread-id",
+            created_at="2023-01-01T00:00:00Z",
+            last_updated_at="2023-01-01T00:01:00Z",
+            tool_name="test_analysis",  # Original tool
+            turns=[
+                ConversationTurn(
+                    role="assistant",
+                    content="Initial analysis",
+                    timestamp="2023-01-01T00:01:00Z",
+                    tool_name="test_analysis",
+                )
+            ],
+            initial_context={"code": "test"},
+        )
+
+        # Mock get_thread to return the existing context
+        mock_get_thread.return_value = existing_context
+
+        # Add turn from review tool
+        from utils.conversation_memory import add_turn
+
+        success = add_turn(
+            "test-thread-id",
+            "assistant",
+            "Review completed",
+            tool_name="test_review",  # Different tool
+        )
+
+        # Verify the add_turn succeeded (basic cross-tool functionality test)
+        assert success
+
+        # Verify thread's original tool_name is preserved
+        setex_calls = mock_client.setex.call_args_list
+        updated_thread_data = setex_calls[-1][0][2]
+        updated_context = json.loads(updated_thread_data)
+
+        assert updated_context["tool_name"] == "test_analysis"  # Original preserved
+        assert len(updated_context["turns"]) == 2
+        assert updated_context["turns"][0]["tool_name"] == "test_analysis"
+        assert updated_context["turns"][1]["tool_name"] == "test_review"
+
+
+if __name__ == "__main__":
+    pytest.main([__file__])
diff --git a/tests/test_thinking_modes.py b/tests/test_thinking_modes.py
index 135072a..c8d441e 100644
--- a/tests/test_thinking_modes.py
+++ b/tests/test_thinking_modes.py
@@ -32,9 +32,9 @@ class TestThinkingModes:
         ]
 
         for tool, expected_default in tools:
-            assert tool.get_default_thinking_mode() == expected_default, (
-                f"{tool.__class__.__name__} should default to {expected_default}"
-            )
+            assert (
+                tool.get_default_thinking_mode() == expected_default
+            ), f"{tool.__class__.__name__} should default to {expected_default}"
 
     @pytest.mark.asyncio
     @patch("tools.base.BaseTool.create_model")
diff --git a/tools/analyze.py b/tools/analyze.py
index 140fe9f..ec9cb65 100644
--- a/tools/analyze.py
+++ b/tools/analyze.py
@@ -90,7 +90,7 @@ class AnalyzeTool(BaseTool):
                 },
                 "continuation_id": {
                     "type": "string",
-                    "description": "Thread continuation ID for multi-turn conversations. Only provide this if continuing a previous conversation thread.",
+                    "description": "Thread continuation ID for multi-turn conversations. Can be used to continue conversations across different tools. Only provide this if continuing a previous conversation thread.",
                 },
             },
             "required": ["files", "question"],
diff --git a/tools/base.py b/tools/base.py
index 796b65b..851f093 100644
--- a/tools/base.py
+++ b/tools/base.py
@@ -14,7 +14,9 @@ Key responsibilities:
 """
 
 import json
+import logging
 import os
+import re
 from abc import ABC, abstractmethod
 from typing import Any, Literal, Optional
 
@@ -23,7 +25,15 @@ from google.genai import types
 from mcp.types import TextContent
 from pydantic import BaseModel, Field
 
-from config import MCP_PROMPT_SIZE_LIMIT
+from config import GEMINI_MODEL, MAX_CONTEXT_TOKENS, MCP_PROMPT_SIZE_LIMIT
+from utils import check_token_limit
+from utils.conversation_memory import (
+    MAX_CONVERSATION_TURNS,
+    add_turn,
+    build_conversation_history,
+    create_thread,
+    get_thread,
+)
 from utils.file_utils import read_file_content, translate_path_for_environment
 
 from .models import ClarificationRequest, ContinuationOffer, FollowUpRequest, ToolOutput
@@ -52,7 +62,7 @@ class ToolRequest(BaseModel):
     )
     continuation_id: Optional[str] = Field(
         None,
-        description="Thread continuation ID for multi-turn conversations. Only provide this if continuing a previous conversation thread.",
+        description="Thread continuation ID for multi-turn conversations. Can be used to continue conversations across different tools. Only provide this if continuing a previous conversation thread.",
     )
 
 
@@ -359,10 +369,15 @@ If any of these would strengthen your analysis, specify what Claude should searc
             List[TextContent]: Formatted response as MCP TextContent objects
         """
         try:
+            # Set up logger for this tool execution
+            logger = logging.getLogger(f"tools.{self.name}")
+            logger.info(f"Starting {self.name} tool execution with arguments: {list(arguments.keys())}")
+
             # Validate request using the tool's Pydantic model
             # This ensures all required fields are present and properly typed
             request_model = self.get_request_model()
             request = request_model(**arguments)
+            logger.debug(f"Request validation successful for {self.name}")
 
             # Validate file paths for security
             # This prevents path traversal attacks and ensures proper access control
@@ -383,13 +398,13 @@ If any of these would strengthen your analysis, specify what Claude should searc
             continuation_id = getattr(request, "continuation_id", None)
             if not continuation_id:
                 # Import here to avoid circular imports
-                import logging
-
                 from server import get_follow_up_instructions
 
                 follow_up_instructions = get_follow_up_instructions(0)  # New conversation, turn 0
                 prompt = f"{prompt}\n\n{follow_up_instructions}"
-                logging.debug(f"Added follow-up instructions for new {self.name} conversation")
+
+                logger.debug(f"Added follow-up instructions for new {self.name} conversation")
+
                 # Also log to file for debugging MCP issues
                 try:
                     with open("/tmp/gemini_debug.log", "a") as f:
@@ -397,13 +412,18 @@ If any of these would strengthen your analysis, specify what Claude should searc
                 except Exception:
                     pass
             else:
-                import logging
+                logger.debug(f"Continuing {self.name} conversation with thread {continuation_id}")
 
-                logging.debug(f"Continuing {self.name} conversation with thread {continuation_id}")
+                # Add conversation history when continuing a threaded conversation
+                thread_context = get_thread(continuation_id)
+                if thread_context:
+                    conversation_history = build_conversation_history(thread_context)
+                    prompt = f"{conversation_history}\n\n{prompt}"
+                    logger.debug(f"Added conversation history to {self.name} prompt for thread {continuation_id}")
+                else:
+                    logger.warning(f"Thread {continuation_id} not found for {self.name} - continuing without history")
 
             # Extract model configuration from request or use defaults
-            from config import GEMINI_MODEL
-
             model_name = getattr(request, "model", None) or GEMINI_MODEL
             temperature = getattr(request, "temperature", None)
             if temperature is None:
@@ -417,7 +437,10 @@ If any of these would strengthen your analysis, specify what Claude should searc
             model = self.create_model(model_name, temperature, thinking_mode)
 
             # Generate AI response using the configured model
+            logger.info(f"Sending request to Gemini API for {self.name}")
+            logger.debug(f"Prompt length: {len(prompt)} characters")
             response = model.generate_content(prompt)
+            logger.info(f"Received response from Gemini API for {self.name}")
 
             # Process the model's response
             if response.candidates and response.candidates[0].content.parts:
@@ -425,11 +448,13 @@ If any of these would strengthen your analysis, specify what Claude should searc
 
                 # Parse response to check for clarification requests or format output
                 tool_output = self._parse_response(raw_text, request)
+                logger.info(f"Successfully completed {self.name} tool execution")
 
             else:
                 # Handle cases where the model couldn't generate a response
                 # This might happen due to safety filters or other constraints
                 finish_reason = response.candidates[0].finish_reason if response.candidates else "Unknown"
+                logger.warning(f"Response blocked or incomplete for {self.name}. Finish reason: {finish_reason}")
                 tool_output = ToolOutput(
                     status="error",
                     content=f"Response blocked or incomplete. Finish reason: {finish_reason}",
@@ -442,6 +467,9 @@ If any of these would strengthen your analysis, specify what Claude should searc
         except Exception as e:
             # Catch all exceptions to prevent server crashes
             # Return error information in standardized format
+            logger = logging.getLogger(f"tools.{self.name}")
+            logger.error(f"Error in {self.name} tool execution: {str(e)}", exc_info=True)
+
             error_output = ToolOutput(
                 status="error",
                 content=f"Error in {self.name}: {str(e)}",
@@ -465,15 +493,14 @@ If any of these would strengthen your analysis, specify what Claude should searc
         """
         # Check for follow-up questions in JSON blocks at the end of the response
         follow_up_question = self._extract_follow_up_question(raw_text)
-
-        import logging
+        logger = logging.getLogger(f"tools.{self.name}")
 
         if follow_up_question:
-            logging.debug(
+            logger.debug(
                 f"Found follow-up question in {self.name} response: {follow_up_question.get('follow_up_question', 'N/A')}"
             )
         else:
-            logging.debug(f"No follow-up question found in {self.name} response")
+            logger.debug(f"No follow-up question found in {self.name} response")
 
         try:
             # Try to parse as JSON to check for clarification requests
@@ -505,15 +532,27 @@ If any of these would strengthen your analysis, specify what Claude should searc
         # Check if we should offer Claude a continuation opportunity
         continuation_offer = self._check_continuation_opportunity(request)
 
-        import logging
-
         if continuation_offer:
-            logging.debug(
+            logger.debug(
                 f"Creating continuation offer for {self.name} with {continuation_offer['remaining_turns']} turns remaining"
             )
             return self._create_continuation_offer_response(formatted_content, continuation_offer, request)
         else:
-            logging.debug(f"No continuation offer created for {self.name}")
+            logger.debug(f"No continuation offer created for {self.name}")
+
+        # If this is a threaded conversation (has continuation_id), save the response
+        continuation_id = getattr(request, "continuation_id", None)
+        if continuation_id:
+            request_files = getattr(request, "files", []) or []
+            success = add_turn(
+                continuation_id,
+                "assistant",
+                formatted_content,
+                files=request_files,
+                tool_name=self.name,
+            )
+            if not success:
+                logging.warning(f"Failed to add turn to thread {continuation_id} for {self.name}")
 
         # Determine content type based on the formatted content
         content_type = (
@@ -539,8 +578,6 @@ If any of these would strengthen your analysis, specify what Claude should searc
         Returns:
             Dict with follow-up data if found, None otherwise
         """
-        import re
-
         # Look for JSON blocks that contain follow_up_question
         # Pattern handles optional leading whitespace and indentation
         json_pattern = r'```json\s*\n\s*(\{.*?"follow_up_question".*?\})\s*\n\s*```'
@@ -573,8 +610,6 @@ If any of these would strengthen your analysis, specify what Claude should searc
         Returns:
             ToolOutput configured for conversation continuation
         """
-        from utils.conversation_memory import add_turn, create_thread
-
         # Create or get thread ID
         continuation_id = getattr(request, "continuation_id", None)
 
@@ -617,9 +652,8 @@ If any of these would strengthen your analysis, specify what Claude should searc
                 )
             except Exception as e:
                 # Threading failed, return normal response
-                import logging
-
-                logging.warning(f"Follow-up threading failed in {self.name}: {str(e)}")
+                logger = logging.getLogger(f"tools.{self.name}")
+                logger.warning(f"Follow-up threading failed in {self.name}: {str(e)}")
                 return ToolOutput(
                     status="success",
                     content=content,
@@ -648,8 +682,6 @@ If any of these would strengthen your analysis, specify what Claude should searc
 
     def _remove_follow_up_json(self, text: str) -> str:
         """Remove follow-up JSON blocks from the response text"""
-        import re
-
         # Remove JSON blocks containing follow_up_question
         pattern = r'```json\s*\n\s*\{.*?"follow_up_question".*?\}\s*\n\s*```'
         return re.sub(pattern, "", text, flags=re.DOTALL).strip()
@@ -676,8 +708,6 @@ If any of these would strengthen your analysis, specify what Claude should searc
 
         # Only offer if we haven't reached conversation limits
         try:
-            from utils.conversation_memory import MAX_CONVERSATION_TURNS
-
             # For new conversations, we have MAX_CONVERSATION_TURNS - 1 remaining
             # (since this response will be turn 1)
             remaining_turns = MAX_CONVERSATION_TURNS - 1
@@ -703,8 +733,6 @@ If any of these would strengthen your analysis, specify what Claude should searc
         Returns:
             ToolOutput configured with continuation offer
         """
-        from utils.conversation_memory import create_thread
-
         try:
             # Create new thread for potential continuation
             thread_id = create_thread(
@@ -712,8 +740,6 @@ If any of these would strengthen your analysis, specify what Claude should searc
             )
 
             # Add this response as the first turn (assistant turn)
-            from utils.conversation_memory import add_turn
-
             request_files = getattr(request, "files", []) or []
             add_turn(thread_id, "assistant", content, files=request_files, tool_name=self.name)
 
@@ -743,9 +769,8 @@ If any of these would strengthen your analysis, specify what Claude should searc
 
         except Exception as e:
             # If threading fails, return normal response but log the error
-            import logging
-
-            logging.warning(f"Conversation threading failed in {self.name}: {str(e)}")
+            logger = logging.getLogger(f"tools.{self.name}")
+            logger.warning(f"Conversation threading failed in {self.name}: {str(e)}")
             return ToolOutput(
                 status="success",
                 content=content,
@@ -800,9 +825,6 @@ If any of these would strengthen your analysis, specify what Claude should searc
         Raises:
             ValueError: If text exceeds MAX_CONTEXT_TOKENS
         """
-        from config import MAX_CONTEXT_TOKENS
-        from utils import check_token_limit
-
         within_limit, estimated_tokens = check_token_limit(text)
         if not within_limit:
             raise ValueError(
diff --git a/tools/chat.py b/tools/chat.py
index e0fbedb..722bc38 100644
--- a/tools/chat.py
+++ b/tools/chat.py
@@ -75,7 +75,7 @@ class ChatTool(BaseTool):
                 },
                 "continuation_id": {
                     "type": "string",
-                    "description": "Thread continuation ID for multi-turn conversations. Only provide this if continuing a previous conversation thread.",
+                    "description": "Thread continuation ID for multi-turn conversations. Can be used to continue conversations across different tools. Only provide this if continuing a previous conversation thread.",
                 },
             },
             "required": ["prompt"],
diff --git a/tools/codereview.py b/tools/codereview.py
index f300367..fa8558c 100644
--- a/tools/codereview.py
+++ b/tools/codereview.py
@@ -128,7 +128,7 @@ class CodeReviewTool(BaseTool):
                 },
                 "continuation_id": {
                     "type": "string",
-                    "description": "Thread continuation ID for multi-turn conversations. Only provide this if continuing a previous conversation thread.",
+                    "description": "Thread continuation ID for multi-turn conversations. Can be used to continue conversations across different tools. Only provide this if continuing a previous conversation thread.",
                 },
             },
             "required": ["files", "context"],
diff --git a/tools/debug.py b/tools/debug.py
index 3d29d08..eedb48f 100644
--- a/tools/debug.py
+++ b/tools/debug.py
@@ -93,7 +93,7 @@ class DebugIssueTool(BaseTool):
                 },
                 "continuation_id": {
                     "type": "string",
-                    "description": "Thread continuation ID for multi-turn conversations. Only provide this if continuing a previous conversation thread.",
+                    "description": "Thread continuation ID for multi-turn conversations. Can be used to continue conversations across different tools. Only provide this if continuing a previous conversation thread.",
                 },
             },
             "required": ["error_description"],
diff --git a/tools/models.py b/tools/models.py
index f08aa43..64ca054 100644
--- a/tools/models.py
+++ b/tools/models.py
@@ -10,7 +10,9 @@ from pydantic import BaseModel, Field
 class FollowUpRequest(BaseModel):
     """Request for follow-up conversation turn"""
 
-    continuation_id: str = Field(..., description="Thread continuation ID for multi-turn conversations")
+    continuation_id: str = Field(
+        ..., description="Thread continuation ID for multi-turn conversations across different tools"
+    )
     question_to_user: str = Field(..., description="Follow-up question to ask Claude")
     suggested_tool_params: Optional[dict[str, Any]] = Field(
         None, description="Suggested parameters for the next tool call"
@@ -23,7 +25,9 @@ class FollowUpRequest(BaseModel):
 class ContinuationOffer(BaseModel):
     """Offer for Claude to continue conversation when Gemini doesn't ask follow-up"""
 
-    continuation_id: str = Field(..., description="Thread continuation ID for multi-turn conversations")
+    continuation_id: str = Field(
+        ..., description="Thread continuation ID for multi-turn conversations across different tools"
+    )
     message_to_user: str = Field(..., description="Message explaining continuation opportunity to Claude")
     suggested_tool_params: Optional[dict[str, Any]] = Field(
         None, description="Suggested parameters for continued tool usage"
diff --git a/tools/precommit.py b/tools/precommit.py
index 8da5556..3d1eb56 100644
--- a/tools/precommit.py
+++ b/tools/precommit.py
@@ -104,7 +104,7 @@ class Precommit(BaseTool):
         if "properties" in schema and "continuation_id" not in schema["properties"]:
             schema["properties"]["continuation_id"] = {
                 "type": "string",
-                "description": "Thread continuation ID for multi-turn conversations. Only provide this if continuing a previous conversation thread.",
+                "description": "Thread continuation ID for multi-turn conversations. Can be used to continue conversations across different tools. Only provide this if continuing a previous conversation thread.",
             }
         return schema
 
diff --git a/tools/thinkdeep.py b/tools/thinkdeep.py
index e307e18..baf0e69 100644
--- a/tools/thinkdeep.py
+++ b/tools/thinkdeep.py
@@ -89,7 +89,7 @@ class ThinkDeepTool(BaseTool):
                 },
                 "continuation_id": {
                     "type": "string",
-                    "description": "Thread continuation ID for multi-turn conversations. Only provide this if continuing a previous conversation thread.",
+                    "description": "Thread continuation ID for multi-turn conversations. Can be used to continue conversations across different tools. Only provide this if continuing a previous conversation thread.",
                 },
             },
             "required": ["current_analysis"],
diff --git a/utils/conversation_memory.py b/utils/conversation_memory.py
index 08aae41..256f5d1 100644
--- a/utils/conversation_memory.py
+++ b/utils/conversation_memory.py
@@ -2,15 +2,48 @@
 Conversation Memory for AI-to-AI Multi-turn Discussions
 
 This module provides conversation persistence and context reconstruction for
-stateless MCP environments. It enables multi-turn conversations between Claude
-and Gemini by storing conversation state in Redis across independent request cycles.
+stateless MCP (Model Context Protocol) environments. It enables multi-turn
+conversations between Claude and Gemini by storing conversation state in Redis
+across independent request cycles.
+
+ARCHITECTURE OVERVIEW:
+The MCP protocol is inherently stateless - each tool request is independent
+with no memory of previous interactions. This module bridges that gap by:
+
+1. Creating persistent conversation threads with unique UUIDs
+2. Storing complete conversation context (turns, files, metadata) in Redis
+3. Reconstructing conversation history when tools are called with continuation_id
+4. Supporting cross-tool continuation - seamlessly switch between different tools
+   while maintaining full conversation context and file references
+
+CROSS-TOOL CONTINUATION:
+A conversation started with one tool (e.g., 'analyze') can be continued with
+any other tool (e.g., 'codereview', 'debug', 'chat') using the same continuation_id.
+The second tool will have access to:
+- All previous conversation turns and responses
+- File context from previous tools (preserved in conversation history)
+- Original thread metadata and timing information
+- Accumulated knowledge from the entire conversation
 
 Key Features:
-- UUID-based conversation thread identification
-- Turn-by-turn conversation history storage
-- Automatic turn limiting to prevent runaway conversations
+- UUID-based conversation thread identification with security validation
+- Turn-by-turn conversation history storage with tool attribution
+- Cross-tool continuation support - switch tools while preserving context
+- File context preservation - files shared in earlier turns remain accessible
+- Automatic turn limiting (5 turns max) to prevent runaway conversations
 - Context reconstruction for stateless request continuity
-- Redis-based persistence with automatic expiration
+- Redis-based persistence with automatic expiration (1 hour TTL)
+- Thread-safe operations for concurrent access
+- Graceful degradation when Redis is unavailable
+
+USAGE EXAMPLE:
+1. Tool A creates thread: create_thread("analyze", request_data) → returns UUID
+2. Tool A adds response: add_turn(UUID, "assistant", response, files=[...], tool_name="analyze")
+3. Tool B continues thread: get_thread(UUID) → retrieves full context
+4. Tool B sees conversation history via build_conversation_history()
+5. Tool B adds its response: add_turn(UUID, "assistant", response, tool_name="codereview")
+
+This enables true AI-to-AI collaboration across the entire tool ecosystem.
 """
 
 import os
@@ -25,7 +58,20 @@ MAX_CONVERSATION_TURNS = 5  # Maximum turns allowed per conversation thread
 
 
 class ConversationTurn(BaseModel):
-    """Single turn in a conversation"""
+    """
+    Single turn in a conversation
+
+    Represents one exchange in the AI-to-AI conversation, tracking both
+    the content and metadata needed for cross-tool continuation.
+
+    Attributes:
+        role: "user" (Claude) or "assistant" (Gemini)
+        content: The actual message content/response
+        timestamp: ISO timestamp when this turn was created
+        follow_up_question: Optional follow-up question from Gemini to Claude
+        files: List of file paths referenced in this specific turn
+        tool_name: Which tool generated this turn (for cross-tool tracking)
+    """
 
     role: str  # "user" or "assistant"
     content: str
@@ -36,18 +82,43 @@ class ConversationTurn(BaseModel):
 
 
 class ThreadContext(BaseModel):
-    """Complete conversation context"""
+    """
+    Complete conversation context for a thread
+
+    Contains all information needed to reconstruct a conversation state
+    across different tools and request cycles. This is the core data
+    structure that enables cross-tool continuation.
+
+    Attributes:
+        thread_id: UUID identifying this conversation thread
+        created_at: ISO timestamp when thread was created
+        last_updated_at: ISO timestamp of last modification
+        tool_name: Name of the tool that initiated this thread
+        turns: List of all conversation turns in chronological order
+        initial_context: Original request data that started the conversation
+    """
 
     thread_id: str
     created_at: str
     last_updated_at: str
-    tool_name: str
+    tool_name: str  # Tool that created this thread (preserved for attribution)
     turns: list[ConversationTurn]
-    initial_context: dict[str, Any]
+    initial_context: dict[str, Any]  # Original request parameters
 
 
 def get_redis_client():
-    """Get Redis client from environment"""
+    """
+    Get Redis client from environment configuration
+
+    Creates a Redis client using the REDIS_URL environment variable.
+    Defaults to localhost:6379/0 if not specified.
+
+    Returns:
+        redis.Redis: Configured Redis client with decode_responses=True
+
+    Raises:
+        ValueError: If redis package is not installed
+    """
     try:
         import redis
 
@@ -58,11 +129,29 @@ def get_redis_client():
 
 
 def create_thread(tool_name: str, initial_request: dict[str, Any]) -> str:
-    """Create new conversation thread and return thread ID"""
+    """
+    Create new conversation thread and return thread ID
+
+    Initializes a new conversation thread for AI-to-AI discussions.
+    This is called when a tool wants to enable follow-up conversations
+    or when Claude explicitly starts a multi-turn interaction.
+
+    Args:
+        tool_name: Name of the tool creating this thread (e.g., "analyze", "chat")
+        initial_request: Original request parameters (will be filtered for serialization)
+
+    Returns:
+        str: UUID thread identifier that can be used for continuation
+
+    Note:
+        - Thread expires after 1 hour (3600 seconds)
+        - Non-serializable parameters are filtered out automatically
+        - Thread can be continued by any tool using the returned UUID
+    """
     thread_id = str(uuid.uuid4())
     now = datetime.now(timezone.utc).isoformat()
 
-    # Filter out non-serializable parameters
+    # Filter out non-serializable parameters to avoid JSON encoding issues
     filtered_context = {
         k: v
         for k, v in initial_request.items()
@@ -73,12 +162,12 @@ def create_thread(tool_name: str, initial_request: dict[str, Any]) -> str:
         thread_id=thread_id,
         created_at=now,
         last_updated_at=now,
-        tool_name=tool_name,
-        turns=[],
+        tool_name=tool_name,  # Track which tool initiated this conversation
+        turns=[],  # Empty initially, turns added via add_turn()
         initial_context=filtered_context,
     )
 
-    # Store in Redis with 1 hour TTL
+    # Store in Redis with 1 hour TTL to prevent indefinite accumulation
     client = get_redis_client()
     key = f"thread:{thread_id}"
     client.setex(key, 3600, context.model_dump_json())
@@ -87,7 +176,25 @@ def create_thread(tool_name: str, initial_request: dict[str, Any]) -> str:
 
 
 def get_thread(thread_id: str) -> Optional[ThreadContext]:
-    """Retrieve thread context from Redis"""
+    """
+    Retrieve thread context from Redis
+
+    Fetches complete conversation context for cross-tool continuation.
+    This is the core function that enables tools to access conversation
+    history from previous interactions.
+
+    Args:
+        thread_id: UUID of the conversation thread
+
+    Returns:
+        ThreadContext: Complete conversation context if found
+        None: If thread doesn't exist, expired, or invalid UUID
+
+    Security:
+        - Validates UUID format to prevent injection attacks
+        - Handles Redis connection failures gracefully
+        - No error information leakage on failure
+    """
     if not thread_id or not _is_valid_uuid(thread_id):
         return None
 
@@ -100,6 +207,7 @@ def get_thread(thread_id: str) -> Optional[ThreadContext]:
             return ThreadContext.model_validate_json(data)
         return None
     except Exception:
+        # Silently handle errors to avoid exposing Redis details
         return None
 
 
@@ -111,47 +219,99 @@ def add_turn(
     files: Optional[list[str]] = None,
     tool_name: Optional[str] = None,
 ) -> bool:
-    """Add turn to existing thread"""
+    """
+    Add turn to existing thread
+
+    Appends a new conversation turn to an existing thread. This is the core
+    function for building conversation history and enabling cross-tool
+    continuation. Each turn preserves the tool that generated it.
+
+    Args:
+        thread_id: UUID of the conversation thread
+        role: "user" (Claude) or "assistant" (Gemini)
+        content: The actual message/response content
+        follow_up_question: Optional follow-up question from Gemini
+        files: Optional list of files referenced in this turn
+        tool_name: Name of the tool adding this turn (for attribution)
+
+    Returns:
+        bool: True if turn was successfully added, False otherwise
+
+    Failure cases:
+        - Thread doesn't exist or expired
+        - Maximum turn limit reached (5 turns)
+        - Redis connection failure
+
+    Note:
+        - Refreshes thread TTL to 1 hour on successful update
+        - Turn limits prevent runaway conversations
+        - File references are preserved for cross-tool access
+    """
     context = get_thread(thread_id)
     if not context:
         return False
 
-    # Check turn limit
+    # Check turn limit to prevent runaway conversations
     if len(context.turns) >= MAX_CONVERSATION_TURNS:
         return False
 
-    # Add new turn
+    # Create new turn with complete metadata
     turn = ConversationTurn(
         role=role,
         content=content,
         timestamp=datetime.now(timezone.utc).isoformat(),
         follow_up_question=follow_up_question,
-        files=files,
-        tool_name=tool_name,
+        files=files,  # Preserved for cross-tool file context
+        tool_name=tool_name,  # Track which tool generated this turn
     )
 
     context.turns.append(turn)
     context.last_updated_at = datetime.now(timezone.utc).isoformat()
 
-    # Save back to Redis
+    # Save back to Redis and refresh TTL
     try:
         client = get_redis_client()
         key = f"thread:{thread_id}"
-        client.setex(key, 3600, context.model_dump_json())  # Refresh TTL
+        client.setex(key, 3600, context.model_dump_json())  # Refresh TTL to 1 hour
         return True
     except Exception:
         return False
 
 
 def build_conversation_history(context: ThreadContext) -> str:
-    """Build formatted conversation history"""
+    """
+    Build formatted conversation history for tool prompts
+
+    Creates a formatted string representation of the conversation history
+    that can be included in tool prompts to provide context. This is the
+    critical function that enables cross-tool continuation by reconstructing
+    the full conversation context.
+
+    Args:
+        context: ThreadContext containing the complete conversation
+
+    Returns:
+        str: Formatted conversation history ready for inclusion in prompts
+        Empty string if no conversation turns exist
+
+    Format:
+        - Header with thread metadata and turn count
+        - Each turn shows: role, tool used, files referenced, content
+        - Files from previous turns are explicitly listed
+        - Clear delimiters for AI parsing
+        - Continuation instruction at end
+
+    Note:
+        This formatted history allows tools to "see" files and context
+        from previous tools, enabling true cross-tool collaboration.
+    """
     if not context.turns:
         return ""
 
     history_parts = [
         "=== CONVERSATION HISTORY ===",
         f"Thread: {context.thread_id}",
-        f"Tool: {context.tool_name}",
+        f"Tool: {context.tool_name}",  # Original tool that started the conversation
         f"Turn {len(context.turns)}/{MAX_CONVERSATION_TURNS}",
         "",
         "Previous exchanges:",
@@ -160,14 +320,14 @@ def build_conversation_history(context: ThreadContext) -> str:
     for i, turn in enumerate(context.turns, 1):
         role_label = "Claude" if turn.role == "user" else "Gemini"
 
-        # Add turn header with tool info if available
+        # Add turn header with tool attribution for cross-tool tracking
         turn_header = f"\n--- Turn {i} ({role_label}"
         if turn.tool_name:
             turn_header += f" using {turn.tool_name}"
         turn_header += ") ---"
         history_parts.append(turn_header)
 
-        # Add files context if present
+        # Add files context if present - critical for cross-tool file access
         if turn.files:
             history_parts.append(f"📁 Files referenced: {', '.join(turn.files)}")
             history_parts.append("")  # Empty line for readability
@@ -187,7 +347,18 @@ def build_conversation_history(context: ThreadContext) -> str:
 
 
 def _is_valid_uuid(val: str) -> bool:
-    """Validate UUID format for security"""
+    """
+    Validate UUID format for security
+
+    Ensures thread IDs are valid UUIDs to prevent injection attacks
+    and malformed requests.
+
+    Args:
+        val: String to validate as UUID
+
+    Returns:
+        bool: True if valid UUID format, False otherwise
+    """
     try:
         uuid.UUID(val)
         return True