diff --git a/Makefile b/Makefile
index ad549c2..a9887c1 100644
--- a/Makefile
+++ b/Makefile
@@ -4,17 +4,18 @@
 
 help:
 	@echo "DeerFlow Development Commands:"
-	@echo "  make check          - Check if all required tools are installed"
-	@echo "  make install        - Install all dependencies (frontend + backend)"
-	@echo "  make dev            - Start all services (frontend + backend + nginx on localhost:2026)"
-	@echo "  make stop           - Stop all running services"
-	@echo "  make clean          - Clean up processes and temporary files"
+	@echo "  make check           - Check if all required tools are installed"
+	@echo "  make install         - Install all dependencies (frontend + backend)"
+	@echo "  make setup-sandbox   - Pre-pull sandbox container image (recommended)"
+	@echo "  make dev             - Start all services (frontend + backend + nginx on localhost:2026)"
+	@echo "  make stop            - Stop all running services"
+	@echo "  make clean           - Clean up processes and temporary files"
 	@echo ""
 	@echo "Docker Development Commands:"
 	@echo "  make docker-init     - Initialize and install dependencies in Docker containers"
 	@echo "  make docker-start    - Start all services in Docker (localhost:2026)"
 	@echo "  make docker-stop     - Stop Docker development services"
-	@echo "  make docker-logs 	  - View Docker development logs"
+	@echo "  make docker-logs     - View Docker development logs"
 	@echo "  make docker-logs-web - View Docker frontend logs"
 	@echo "  make docker-logs-api - View Docker backend logs"
 
@@ -100,6 +101,43 @@ install:
 	@echo "Installing frontend dependencies..."
 	@cd frontend && pnpm install
 	@echo "✓ All dependencies installed"
+	@echo ""
+	@echo "=========================================="
+	@echo "  Optional: Pre-pull Sandbox Image"
+	@echo "=========================================="
+	@echo ""
+	@echo "If you plan to use Docker/Container-based sandbox, you can pre-pull the image:"
+	@echo "  make setup-sandbox"
+	@echo ""
+
+# Pre-pull sandbox Docker image (optional but recommended)
+setup-sandbox:
+	@echo "=========================================="
+	@echo "  Pre-pulling Sandbox Container Image"
+	@echo "=========================================="
+	@echo ""
+	@IMAGE=$$(grep -A 20 "# sandbox:" config.yaml 2>/dev/null | grep "image:" | awk '{print $$2}' | head -1); \
+	if [ -z "$$IMAGE" ]; then \
+		IMAGE="enterprise-public-cn-beijing.cr.volces.com/vefaas-public/all-in-one-sandbox:latest"; \
+		echo "Using default image: $$IMAGE"; \
+	else \
+		echo "Using configured image: $$IMAGE"; \
+	fi; \
+	echo ""; \
+	if command -v container >/dev/null 2>&1 && [ "$$(uname)" = "Darwin" ]; then \
+		echo "Detected Apple Container on macOS, pulling image..."; \
+		container pull "$$IMAGE" || echo "⚠ Apple Container pull failed, will try Docker"; \
+	fi; \
+	if command -v docker >/dev/null 2>&1; then \
+		echo "Pulling image using Docker..."; \
+		docker pull "$$IMAGE"; \
+		echo ""; \
+		echo "✓ Sandbox image pulled successfully"; \
+	else \
+		echo "✗ Neither Docker nor Apple Container is available"; \
+		echo "  Please install Docker: https://docs.docker.com/get-docker/"; \
+		exit 1; \
+	fi
 
 # Start all services
 dev:
@@ -110,7 +148,7 @@ dev:
 	@-nginx -c $(PWD)/docker/nginx/nginx.local.conf -p $(PWD) -s quit 2>/dev/null || true
 	@sleep 1
 	@-pkill -9 nginx 2>/dev/null || true
-	@-docker ps -q --filter "name=deer-flow-sandbox" | xargs -r docker stop 2>/dev/null || true
+	@-./scripts/cleanup-containers.sh deer-flow-sandbox 2>/dev/null || true
 	@sleep 1
 	@echo ""
 	@echo "=========================================="
@@ -132,7 +170,7 @@ dev:
 		sleep 1; \
 		pkill -9 nginx 2>/dev/null || true; \
 		echo "Cleaning up sandbox containers..."; \
-		docker ps -q --filter "name=deer-flow-sandbox" | xargs -r docker stop 2>/dev/null || true; \
+		./scripts/cleanup-containers.sh deer-flow-sandbox 2>/dev/null || true; \
 		echo "✓ All services stopped"; \
 		exit 0; \
 	}; \
@@ -183,7 +221,7 @@ stop:
 	@sleep 1
 	@-pkill -9 nginx 2>/dev/null || true
 	@echo "Cleaning up sandbox containers..."
-	@-docker ps -q --filter "name=deer-flow-sandbox" | xargs -r docker stop 2>/dev/null || true
+	@-./scripts/cleanup-containers.sh deer-flow-sandbox 2>/dev/null || true
 	@echo "✓ All services stopped"
 
 # Clean up
diff --git a/README.md b/README.md
index 288b539..2273c0b 100644
--- a/README.md
+++ b/README.md
@@ -41,18 +41,25 @@ If you prefer running services locally:
    make install
    ```
 
-3. **Start services**:
+3. **(Optional) Pre-pull sandbox image**:
+   ```bash
+   # Recommended if using Docker/Container-based sandbox
+   make setup-sandbox
+   ```
+
+4. **Start services**:
    ```bash
    make dev
    ```
 
-4. **Access**: http://localhost:2026
+5. **Access**: http://localhost:2026
 
 See [CONTRIBUTING.md](CONTRIBUTING.md) for detailed local development guide.
 
 ## Features
 
 - 🤖 **LangGraph-based Agents** - Multi-agent orchestration with sophisticated workflows
+- 🧠 **Persistent Memory** - LLM-powered context retention across conversations with automatic fact extraction
 - 🔧 **Model Context Protocol (MCP)** - Extensible tool integration
 - 🎯 **Skills System** - Reusable agent capabilities
 - 🛡️ **Sandbox Execution** - Safe code execution environment
diff --git a/backend/CLAUDE.md b/backend/CLAUDE.md
index 1d48aad..d8c56c6 100644
--- a/backend/CLAUDE.md
+++ b/backend/CLAUDE.md
@@ -40,6 +40,17 @@ deer-flow/
     └── custom/                # Custom skills (gitignored)
 ```
 
+## Important Development Guidelines
+
+### Documentation Update Policy
+**CRITICAL: Always update README.md and CLAUDE.md after every code change**
+
+When making code changes, you MUST update the relevant documentation:
+- Update `README.md` for user-facing changes (features, setup, usage instructions)
+- Update `CLAUDE.md` for development changes (architecture, commands, workflows, internal systems)
+- Keep documentation synchronized with the codebase at all times
+- Ensure accuracy and timeliness of all documentation
+
 ## Commands
 
 **Root directory** (for full application):
@@ -202,7 +213,49 @@ Configuration priority:
   5. `TitleMiddleware` - Generates conversation titles
   6. `TodoListMiddleware` - Tracks multi-step tasks (if plan_mode enabled)
   7. `ViewImageMiddleware` - Injects image details for vision models
-  8. `ClarificationMiddleware` - Handles clarification requests (must be last)
+  8. `MemoryMiddleware` - Automatic context retention and personalization (if enabled)
+  9. `ClarificationMiddleware` - Handles clarification requests (must be last)
+
+**Memory System** (`src/agents/memory/`)
+- LLM-powered personalization layer that automatically extracts and stores user context across conversations
+- Components:
+  - `updater.py` - LLM-based memory updates with fact extraction and file I/O
+  - `queue.py` - Debounced update queue for batching and performance optimization
+  - `prompt.py` - Prompt templates and formatting utilities for memory updates
+- `MemoryMiddleware` (`src/agents/middlewares/memory_middleware.py`) - Queues conversations for memory updates
+- Gateway API (`src/gateway/routers/memory.py`) - REST endpoints for memory management
+- Storage: JSON file at `backend/.deer-flow/memory.json`
+
+**Memory Data Structure**:
+- **User Context** (current state):
+  - `workContext` - Work-related information (job, projects, technologies)
+  - `personalContext` - Preferences, communication style, background
+  - `topOfMind` - Current focus areas and immediate priorities
+- **History** (temporal context):
+  - `recentMonths` - Recent activities and discussions
+  - `earlierContext` - Important historical context
+  - `longTermBackground` - Persistent background information
+- **Facts** (structured knowledge):
+  - Discrete facts with categories: `preference`, `knowledge`, `context`, `behavior`, `goal`
+  - Each fact includes: `id`, `content`, `category`, `confidence` (0-1), `createdAt`, `source` (thread ID)
+  - Confidence threshold (default 0.7) filters low-quality facts
+  - Max facts limit (default 100) keeps highest-confidence facts
+
+**Memory Workflow**:
+1. **Post-Interaction**: `MemoryMiddleware` filters messages (user inputs + final AI responses only) and queues conversation
+2. **Debounced Processing**: Queue waits 30s (configurable), batches multiple updates, resets timer on new updates
+3. **LLM-Based Update**: Background thread loads memory, formats conversation, invokes LLM to extract:
+   - Updated context summaries (1-3 sentences each)
+   - New facts with confidence scores and categories
+   - Facts to remove (contradictions)
+4. **Storage**: Applies updates atomically to `memory.json` with cache invalidation (mtime-based)
+5. **Injection**: Next interaction loads memory, formats top 15 facts + context, injects into `<memory>` tags in system prompt
+
+**Memory API Endpoints** (`/api/memory`):
+- `GET /api/memory` - Retrieve current memory data
+- `POST /api/memory/reload` - Force reload from file (invalidates cache)
+- `GET /api/memory/config` - Get memory configuration
+- `GET /api/memory/status` - Get both config and data
 
 ### Config Schema
 
@@ -215,6 +268,15 @@ Models, tools, sandbox providers, skills, and middleware settings are configured
 - `skills.container_path`: Container mount path (default: `/mnt/skills`)
 - `title`: Automatic thread title generation configuration
 - `summarization`: Automatic conversation summarization configuration
+- `memory`: Memory system configuration
+  - `enabled`: Master switch (boolean)
+  - `storage_path`: Path to memory.json file (relative to backend/)
+  - `debounce_seconds`: Wait time before processing updates (default: 30)
+  - `model_name`: LLM model for memory updates (null = use default model)
+  - `max_facts`: Maximum facts to store (default: 100)
+  - `fact_confidence_threshold`: Minimum confidence to store fact (default: 0.7)
+  - `injection_enabled`: Inject memory into system prompt (boolean)
+  - `max_injection_tokens`: Token limit for memory injection (default: 2000)
 
 **Extensions Configuration Schema** (`extensions_config.json`):
 - `mcpServers`: Map of MCP server name to configuration
@@ -307,6 +369,29 @@ For models with `supports_vision: true`:
 - `view_image_tool` added to agent's toolset
 - Images automatically converted and injected into state
 
+### Memory System
+
+Persistent context retention and personalization across conversations:
+- **Automatic Extraction**: LLM analyzes conversations to extract user context, facts, and preferences
+- **Structured Storage**: Maintains user context, history, and confidence-scored facts in JSON format
+- **Smart Filtering**: Only processes meaningful messages (user inputs + final AI responses)
+- **Debounced Updates**: Batches updates to minimize LLM calls (configurable wait time)
+- **System Prompt Injection**: Automatically injects relevant memory context into agent prompts
+- **Cache Optimization**: File modification time-based cache invalidation for external edits
+- **Thread Safety**: Locks protect queue and cache for concurrent access
+- **REST API**: Full CRUD operations via `/api/memory` endpoints
+- **Frontend Integration**: Memory settings page for viewing and managing memory data
+
+**Configuration**: Controlled via `memory` section in `config.yaml`
+- Enable/disable memory system
+- Configure storage path, debounce timing, fact limits
+- Control system prompt injection and token limits
+- Set confidence thresholds for fact storage
+
+**Storage Location**: `backend/.deer-flow/memory.json`
+
+See configuration section for detailed settings.
+
 ## Code Style
 
 - Uses `ruff` for linting and formatting
diff --git a/backend/docs/APPLE_CONTAINER.md b/backend/docs/APPLE_CONTAINER.md
new file mode 100644
index 0000000..6ef82d0
--- /dev/null
+++ b/backend/docs/APPLE_CONTAINER.md
@@ -0,0 +1,238 @@
+# Apple Container Support
+
+DeerFlow now supports Apple Container as the preferred container runtime on macOS, with automatic fallback to Docker.
+
+## Overview
+
+Starting with this version, DeerFlow automatically detects and uses Apple Container on macOS when available, falling back to Docker when:
+- Apple Container is not installed
+- Running on non-macOS platforms
+
+This provides better performance on Apple Silicon Macs while maintaining compatibility across all platforms.
+
+## Benefits
+
+### On Apple Silicon Macs with Apple Container:
+- **Better Performance**: Native ARM64 execution without Rosetta 2 translation
+- **Lower Resource Usage**: Lighter weight than Docker Desktop
+- **Native Integration**: Uses macOS Virtualization.framework
+
+### Fallback to Docker:
+- Full backward compatibility
+- Works on all platforms (macOS, Linux, Windows)
+- No configuration changes needed
+
+## Requirements
+
+### For Apple Container (macOS only):
+- macOS 15.0 or later
+- Apple Silicon (M1/M2/M3/M4)
+- Apple Container CLI installed
+
+### Installation:
+```bash
+# Download from GitHub releases
+# https://github.com/apple/container/releases
+
+# Verify installation
+container --version
+
+# Start the service
+container system start
+```
+
+### For Docker (all platforms):
+- Docker Desktop or Docker Engine
+
+## How It Works
+
+### Automatic Detection
+
+The `AioSandboxProvider` automatically detects the available container runtime:
+
+1. On macOS: Try `container --version`
+   - Success → Use Apple Container
+   - Failure → Fall back to Docker
+
+2. On other platforms: Use Docker directly
+
+### Runtime Differences
+
+Both runtimes use nearly identical command syntax:
+
+**Container Startup:**
+```bash
+# Apple Container
+container run --rm -d -p 8080:8080 -v /host:/container -e KEY=value image
+
+# Docker
+docker run --rm -d -p 8080:8080 -v /host:/container -e KEY=value image
+```
+
+**Container Cleanup:**
+```bash
+# Apple Container (with --rm flag)
+container stop <id>  # Auto-removes due to --rm
+
+# Docker (with --rm flag)
+docker stop <id>     # Auto-removes due to --rm
+```
+
+### Implementation Details
+
+The implementation is in `backend/src/community/aio_sandbox/aio_sandbox_provider.py`:
+
+- `_detect_container_runtime()`: Detects available runtime at startup
+- `_start_container()`: Uses detected runtime, skips Docker-specific options for Apple Container
+- `_stop_container()`: Uses appropriate stop command for the runtime
+
+## Configuration
+
+No configuration changes are needed! The system works automatically.
+
+However, you can verify the runtime in use by checking the logs:
+
+```
+INFO:src.community.aio_sandbox.aio_sandbox_provider:Detected Apple Container: container version 0.1.0
+INFO:src.community.aio_sandbox.aio_sandbox_provider:Starting sandbox container using container: ...
+```
+
+Or for Docker:
+```
+INFO:src.community.aio_sandbox.aio_sandbox_provider:Apple Container not available, falling back to Docker
+INFO:src.community.aio_sandbox.aio_sandbox_provider:Starting sandbox container using docker: ...
+```
+
+## Container Images
+
+Both runtimes use OCI-compatible images. The default image works with both:
+
+```yaml
+sandbox:
+  use: src.community.aio_sandbox:AioSandboxProvider
+  image: enterprise-public-cn-beijing.cr.volces.com/vefaas-public/all-in-one-sandbox:latest  # Default image
+```
+
+Make sure your images are available for the appropriate architecture:
+- ARM64 for Apple Container on Apple Silicon
+- AMD64 for Docker on Intel Macs
+- Multi-arch images work on both
+
+### Pre-pulling Images (Recommended)
+
+**Important**: Container images are typically large (500MB+) and are pulled on first use, which can cause a long wait time without clear feedback.
+
+**Best Practice**: Pre-pull the image during setup:
+
+```bash
+# From project root
+make setup-sandbox
+```
+
+This command will:
+1. Read the configured image from `config.yaml` (or use default)
+2. Detect available runtime (Apple Container or Docker)
+3. Pull the image with progress indication
+4. Verify the image is ready for use
+
+**Manual pre-pull**:
+
+```bash
+# Using Apple Container
+container pull enterprise-public-cn-beijing.cr.volces.com/vefaas-public/all-in-one-sandbox:latest
+
+# Using Docker
+docker pull enterprise-public-cn-beijing.cr.volces.com/vefaas-public/all-in-one-sandbox:latest
+```
+
+If you skip pre-pulling, the image will be automatically pulled on first agent execution, which may take several minutes depending on your network speed.
+
+## Cleanup Scripts
+
+The project includes a unified cleanup script that handles both runtimes:
+
+**Script:** `scripts/cleanup-containers.sh`
+
+**Usage:**
+```bash
+# Clean up all DeerFlow sandbox containers
+./scripts/cleanup-containers.sh deer-flow-sandbox
+
+# Custom prefix
+./scripts/cleanup-containers.sh my-prefix
+```
+
+**Makefile Integration:**
+
+All cleanup commands in `Makefile` automatically handle both runtimes:
+```bash
+make stop   # Stops all services and cleans up containers
+make clean  # Full cleanup including logs
+```
+
+## Testing
+
+Test the container runtime detection:
+
+```bash
+cd backend
+python test_container_runtime.py
+```
+
+This will:
+1. Detect the available runtime
+2. Optionally start a test container
+3. Verify connectivity
+4. Clean up
+
+## Troubleshooting
+
+### Apple Container not detected on macOS
+
+1. Check if installed:
+   ```bash
+   which container
+   container --version
+   ```
+
+2. Check if service is running:
+   ```bash
+   container system start
+   ```
+
+3. Check logs for detection:
+   ```bash
+   # Look for detection message in application logs
+   grep "container runtime" logs/*.log
+   ```
+
+### Containers not cleaning up
+
+1. Manually check running containers:
+   ```bash
+   # Apple Container
+   container list
+
+   # Docker
+   docker ps
+   ```
+
+2. Run cleanup script manually:
+   ```bash
+   ./scripts/cleanup-containers.sh deer-flow-sandbox
+   ```
+
+### Performance issues
+
+- Apple Container should be faster on Apple Silicon
+- If experiencing issues, you can force Docker by temporarily renaming the `container` command:
+   ```bash
+   # Temporary workaround - not recommended for permanent use
+   sudo mv /opt/homebrew/bin/container /opt/homebrew/bin/container.bak
+   ```
+
+## References
+
+- [Apple Container GitHub](https://github.com/apple/container)
+- [Apple Container Documentation](https://github.com/apple/container/blob/main/docs/)
+- [OCI Image Spec](https://github.com/opencontainers/image-spec)
diff --git a/backend/docs/SETUP.md b/backend/docs/SETUP.md
index 411268b..9e9214f 100644
--- a/backend/docs/SETUP.md
+++ b/backend/docs/SETUP.md
@@ -49,6 +49,22 @@ The backend searches for `config.yaml` in this order:
 
 **Recommended**: Place `config.yaml` in project root (`deer-flow/config.yaml`).
 
+## Sandbox Setup (Optional but Recommended)
+
+If you plan to use Docker/Container-based sandbox (configured in `config.yaml` under `sandbox.use: src.community.aio_sandbox:AioSandboxProvider`), it's highly recommended to pre-pull the container image:
+
+```bash
+# From project root
+make setup-sandbox
+```
+
+**Why pre-pull?**
+- The sandbox image (~500MB+) is pulled on first use, causing a long wait
+- Pre-pulling provides clear progress indication
+- Avoids confusion when first using the agent
+
+If you skip this step, the image will be automatically pulled on first agent execution, which may take several minutes depending on your network speed.
+
 ## Troubleshooting
 
 ### Config file not found
diff --git a/backend/src/community/aio_sandbox/aio_sandbox_provider.py b/backend/src/community/aio_sandbox/aio_sandbox_provider.py
index c47caaf..8edb36b 100644
--- a/backend/src/community/aio_sandbox/aio_sandbox_provider.py
+++ b/backend/src/community/aio_sandbox/aio_sandbox_provider.py
@@ -32,14 +32,17 @@ IDLE_CHECK_INTERVAL = 60  # Check every 60 seconds
 
 
 class AioSandboxProvider(SandboxProvider):
-    """Sandbox provider that manages Docker containers running the AIO sandbox.
+    """Sandbox provider that manages containers running the AIO sandbox.
+
+    On macOS, automatically prefers Apple Container if available, otherwise falls back to Docker.
+    On other platforms, uses Docker.
 
     Configuration options in config.yaml under sandbox:
         use: src.community.aio_sandbox:AioSandboxProvider
-        image: enterprise-public-cn-beijing.cr.volces.com/vefaas-public/all-in-one-sandbox:latest  # Docker image to use
+        image: enterprise-public-cn-beijing.cr.volces.com/vefaas-public/all-in-one-sandbox:latest  # Container image to use (works with both runtimes)
         port: 8080  # Base port for sandbox containers
         base_url: http://localhost:8080  # If set, uses existing sandbox instead of starting new container
-        auto_start: true  # Whether to automatically start Docker container
+        auto_start: true  # Whether to automatically start container
         container_prefix: deer-flow-sandbox  # Prefix for container names
         idle_timeout: 600  # Idle timeout in seconds (default: 600 = 10 minutes). Set to 0 to disable.
         mounts:  # List of volume mounts
@@ -62,6 +65,7 @@ class AioSandboxProvider(SandboxProvider):
         self._shutdown_called = False
         self._idle_checker_stop = threading.Event()
         self._idle_checker_thread: threading.Thread | None = None
+        self._container_runtime = self._detect_container_runtime()
 
         # Register shutdown handler to clean up containers on exit
         atexit.register(self.shutdown)
@@ -184,6 +188,35 @@ class AioSandboxProvider(SandboxProvider):
                 resolved[key] = str(value)
         return resolved
 
+    def _detect_container_runtime(self) -> str:
+        """Detect which container runtime to use.
+
+        On macOS, prefer Apple Container if available, otherwise fall back to Docker.
+        On other platforms, use Docker.
+
+        Returns:
+            "container" for Apple Container, "docker" for Docker.
+        """
+        import platform
+
+        # Only try Apple Container on macOS
+        if platform.system() == "Darwin":
+            try:
+                result = subprocess.run(
+                    ["container", "--version"],
+                    capture_output=True,
+                    text=True,
+                    check=True,
+                    timeout=5,
+                )
+                logger.info(f"Detected Apple Container: {result.stdout.strip()}")
+                return "container"
+            except (FileNotFoundError, subprocess.CalledProcessError, subprocess.TimeoutExpired):
+                logger.info("Apple Container not available, falling back to Docker")
+
+        # Default to Docker
+        return "docker"
+
     def _is_sandbox_ready(self, base_url: str, timeout: int = 30) -> bool:
         """Check if sandbox is ready to accept connections.
 
@@ -253,7 +286,10 @@ class AioSandboxProvider(SandboxProvider):
         return None
 
     def _start_container(self, sandbox_id: str, port: int, extra_mounts: list[tuple[str, str, bool]] | None = None) -> str:
-        """Start a new Docker container for the sandbox.
+        """Start a new container for the sandbox.
+
+        On macOS, prefers Apple Container if available, otherwise uses Docker.
+        On other platforms, uses Docker.
 
         Args:
             sandbox_id: Unique identifier for the sandbox.
@@ -267,17 +303,22 @@ class AioSandboxProvider(SandboxProvider):
         container_name = f"{self._config['container_prefix']}-{sandbox_id}"
 
         cmd = [
-            "docker",
+            self._container_runtime,
             "run",
-            "--security-opt",
-            "seccomp=unconfined",
+        ]
+
+        # Add Docker-specific security options
+        if self._container_runtime == "docker":
+            cmd.extend(["--security-opt", "seccomp=unconfined"])
+
+        cmd.extend([
             "--rm",
             "-d",
             "-p",
             f"{port}:8080",
             "--name",
             container_name,
-        ]
+        ])
 
         # Add configured environment variables
         for key, value in self._config["environment"].items():
@@ -303,26 +344,28 @@ class AioSandboxProvider(SandboxProvider):
 
         cmd.append(image)
 
-        logger.info(f"Starting sandbox container: {' '.join(cmd)}")
+        logger.info(f"Starting sandbox container using {self._container_runtime}: {' '.join(cmd)}")
 
         try:
             result = subprocess.run(cmd, capture_output=True, text=True, check=True)
             container_id = result.stdout.strip()
-            logger.info(f"Started sandbox container {container_name} with ID {container_id}")
+            logger.info(f"Started sandbox container {container_name} with ID {container_id} using {self._container_runtime}")
             return container_id
         except subprocess.CalledProcessError as e:
-            logger.error(f"Failed to start sandbox container: {e.stderr}")
+            logger.error(f"Failed to start sandbox container using {self._container_runtime}: {e.stderr}")
             raise RuntimeError(f"Failed to start sandbox container: {e.stderr}")
 
     def _stop_container(self, container_id: str) -> None:
-        """Stop and remove a Docker container.
+        """Stop and remove a container.
+
+        Since we use --rm flag, the container is automatically removed after stopping.
 
         Args:
             container_id: The container ID to stop.
         """
         try:
-            subprocess.run(["docker", "stop", container_id], capture_output=True, text=True, check=True)
-            logger.info(f"Stopped sandbox container {container_id}")
+            subprocess.run([self._container_runtime, "stop", container_id], capture_output=True, text=True, check=True)
+            logger.info(f"Stopped sandbox container {container_id} using {self._container_runtime} (--rm will auto-remove)")
         except subprocess.CalledProcessError as e:
             logger.warning(f"Failed to stop sandbox container {container_id}: {e.stderr}")
 
diff --git a/config.example.yaml b/config.example.yaml
index f1d62b1..3c339d0 100644
--- a/config.example.yaml
+++ b/config.example.yaml
@@ -144,17 +144,20 @@ tools:
 sandbox:
   use: src.sandbox.local:LocalSandboxProvider
 
-# Option 2: Docker-based AIO Sandbox
-# Executes commands in isolated Docker containers
+# Option 2: Container-based AIO Sandbox
+# Executes commands in isolated containers (Docker or Apple Container)
+# On macOS: Automatically prefers Apple Container if available, falls back to Docker
+# On other platforms: Uses Docker
 # Uncomment to use:
 # sandbox:
 #   use: src.community.aio_sandbox:AioSandboxProvider
 #
-#   # Optional: Use existing sandbox at this URL (no Docker container will be started)
+#   # Optional: Use existing sandbox at this URL (no container will be started)
 #   # base_url: http://localhost:8080
 #
-#   # Optional: Docker image to use
+#   # Optional: Container image to use (works with both Docker and Apple Container)
 #   # Default: enterprise-public-cn-beijing.cr.volces.com/vefaas-public/all-in-one-sandbox:latest
+#   # Recommended: enterprise-public-cn-beijing.cr.volces.com/vefaas-public/all-in-one-sandbox:latest (works on both x86_64 and arm64)
 #   # image: enterprise-public-cn-beijing.cr.volces.com/vefaas-public/all-in-one-sandbox:latest
 #
 #   # Optional: Base port for sandbox containers (default: 8080)
diff --git a/scripts/cleanup-containers.sh b/scripts/cleanup-containers.sh
new file mode 100755
index 0000000..7d69a0b
--- /dev/null
+++ b/scripts/cleanup-containers.sh
@@ -0,0 +1,96 @@
+#!/usr/bin/env bash
+#
+# cleanup-containers.sh - Clean up DeerFlow sandbox containers
+#
+# This script cleans up both Docker and Apple Container runtime containers
+# to ensure compatibility across different container runtimes.
+#
+
+set -e
+
+PREFIX="${1:-deer-flow-sandbox}"
+
+# Colors for output
+RED='\033[0;31m'
+GREEN='\033[0;32m'
+YELLOW='\033[1;33m'
+NC='\033[0m' # No Color
+
+echo "Cleaning up sandbox containers with prefix: ${PREFIX}"
+
+# Function to clean up Docker containers
+cleanup_docker() {
+    if command -v docker &> /dev/null; then
+        echo -n "Checking Docker containers... "
+        DOCKER_CONTAINERS=$(docker ps -q --filter "name=${PREFIX}" 2>/dev/null || echo "")
+
+        if [ -n "$DOCKER_CONTAINERS" ]; then
+            echo ""
+            echo "Found Docker containers to clean up:"
+            docker ps --filter "name=${PREFIX}" --format "table {{.ID}}\t{{.Names}}\t{{.Status}}"
+            echo "Stopping Docker containers..."
+            echo "$DOCKER_CONTAINERS" | xargs docker stop 2>/dev/null || true
+            echo -e "${GREEN}✓ Docker containers stopped${NC}"
+        else
+            echo -e "${GREEN}none found${NC}"
+        fi
+    else
+        echo "Docker not found, skipping..."
+    fi
+}
+
+# Function to clean up Apple Container containers
+cleanup_apple_container() {
+    if command -v container &> /dev/null; then
+        echo -n "Checking Apple Container containers... "
+
+        # List all containers and filter by name
+        CONTAINER_LIST=$(container list --format json 2>/dev/null || echo "[]")
+
+        if [ "$CONTAINER_LIST" != "[]" ] && [ -n "$CONTAINER_LIST" ]; then
+            # Extract container IDs that match our prefix
+            CONTAINER_IDS=$(echo "$CONTAINER_LIST" | python3 -c "
+import json
+import sys
+try:
+    containers = json.load(sys.stdin)
+    if isinstance(containers, list):
+        for c in containers:
+            if isinstance(c, dict):
+                name = c.get('name', '')
+                cid = c.get('id', '')
+                if '${PREFIX}' in name and cid:
+                    print(cid)
+except:
+    pass
+" 2>/dev/null || echo "")
+
+            if [ -n "$CONTAINER_IDS" ]; then
+                echo ""
+                echo "Found Apple Container containers to clean up:"
+                echo "$CONTAINER_IDS" | while read -r cid; do
+                    echo "  - $cid"
+                done
+
+                echo "Stopping Apple Container containers..."
+                echo "$CONTAINER_IDS" | while read -r cid; do
+                    container stop "$cid" 2>/dev/null || true
+                    container delete "$cid" 2>/dev/null || true
+                done
+                echo -e "${GREEN}✓ Apple Container containers stopped${NC}"
+            else
+                echo -e "${GREEN}none found${NC}"
+            fi
+        else
+            echo -e "${GREEN}none found${NC}"
+        fi
+    else
+        echo "Apple Container not found, skipping..."
+    fi
+}
+
+# Clean up both runtimes
+cleanup_docker
+cleanup_apple_container
+
+echo -e "${GREEN}✓ Container cleanup complete${NC}"