Release v0.3.0: Tasmota power display, device detection, and SSH fixes

- Added Tasmota power state and consumption display in device list
- Added automatic Tasmota and test equipment detection
- Added device cache system for faster discovery
- Added SSH status display with detailed error information
- Fixed Tasmota detection bug (timeout parameter)
- Fixed SSH authentication and credential caching
- Fixed cache race conditions and error handling
- Improved device type detection and identification
- Optimized parallel device discovery

Author: Alex J Lennon

CHANGELOG.md

@@ -2,6 +2,33 @@
 
 [Semantic Versioning](https://semver.org/)
 
+## [0.3.0] - 2025-11-17
+
+### Added
+- **Tasmota Power State Display**: Tasmota devices now show power state (ON/OFF) and power consumption (Watts) in the device list
+- **Enhanced Device Detection**: Automatic detection of Tasmota devices and test equipment (DMMs) via network protocols
+- **Device Type Detection**: Improved device type inference from hostname patterns (eink, sentai boards)
+- **SSH Status Display**: Device list now shows detailed SSH connection status (OK, Timeout, Refused, Auth Failed)
+- **Firmware Version Display**: Device list displays firmware version information from `/etc/os-release`
+- **VPN Discovery Indicator**: Devices discovered over VPN are marked with a VPN indicator
+- **Device Cache System**: JSON-based caching of device information (hostname, unique ID, firmware) to reduce repeated SSH queries
+- **Friendly Name Management**: Ability to update friendly names for discovered devices via `update_device_friendly_name` tool
+- **Parallel Device Identification**: Optimized device discovery with parallel SSH identification attempts
+
+### Fixed
+- **Tasmota Detection Bug**: Fixed timeout parameter in Tasmota HTTP API detection (was incorrectly passed to Request instead of urlopen)
+- **SSH Authentication**: Fixed credential caching to use correct function signature (username/password parameters)
+- **SSH Error Handling**: Fixed cache to re-verify devices with SSH errors but no hostname instead of returning stale errors
+- **Cache Race Conditions**: Fixed cache merge logic to preserve successful identifications and prevent failures from overwriting successes
+- **Host Key Verification**: Improved SSH host key handling with `StrictHostKeyChecking=accept-new` for new/changed host keys
+- **Device Type Detection**: Fixed device type detection to properly use cached Tasmota/test equipment detection results
+- **Cache Persistence**: Tasmota and test equipment detection results are now properly saved to persistent cache
+
+### Changed
+- **Device List Format**: Device list now displays as a markdown table with improved readability
+- **Device Discovery**: Optimized device discovery to check all devices for Tasmota/test equipment detection, not just uncached ones
+- **SSH Credential Priority**: SSH authentication now prioritizes "fio" user over "root" for device identification
+
 ## [0.2.0] - 2025-11-16
 
 ### Changed

README.md

@@ -7,7 +7,7 @@
 
 MCP server exposing remote embedded hardware testing capabilities to AI assistants.
 
-**Version**: 0.2.0
+**Version**: 0.3.0
 
 > **⚠️ ALPHA QUALITY WARNING**: This package is currently in **alpha** development status. It is **not ready for professional or production use**. The API may change, features may be incomplete, and there may be bugs. Use at your own risk. See [PUBLISHING.md](PUBLISHING.md) for more details.
 
@@ -41,6 +41,8 @@ python3.10 lab_testing/test_server.py
 
 ## Configuration
 
+### MCP Server Configuration
+
 Add to Cursor MCP config (`~/.cursor/mcp.json`):
 
 ```json
@@ -60,6 +62,46 @@ Add to Cursor MCP config (`~/.cursor/mcp.json`):
 
 **Important:** Use `python3.10` (or `python3.11+`) since MCP SDK requires Python 3.10+.
 
+### Target Network Configuration
+
+Configure the target network for lab testing operations:
+
+**Option 1: Environment Variable** (in `~/.cursor/mcp.json`):
+```json
+{
+  "mcpServers": {
+    "ai-lab-testing": {
+      "command": "python3.10",
+      "args": ["/path/to/ai-lab-testing/lab_testing/server.py"],
+      "env": {
+        "LAB_TESTING_ROOT": "/path/to/ai-lab-testing",
+        "TARGET_NETWORK": "192.168.2.0/24"
+      }
+    }
+  }
+}
+```
+
+**Option 2: Config File** (in `lab_devices.json`):
+```json
+{
+  "lab_infrastructure": {
+    "network_access": {
+      "target_network": "192.168.2.0/24",
+      "lab_networks": ["192.168.2.0/24"]
+    }
+  }
+}
+```
+
+**Default**: `192.168.2.0/24` if not configured.
+
+### Tool Timeouts
+
+Some tools like `create_network_map` may take longer than 30 seconds for full network scans. If tool calls timeout:
+- Use `quick_mode: true` to skip network scanning (completes in <5s)
+- Network scanning has been optimized with faster timeouts and increased parallelism
+
 ### VPN Setup
 
 The server auto-detects WireGuard VPN configs. If you don't have one:

docs/SETUP.md

@@ -27,6 +27,33 @@ Uses existing lab testing framework:
 
 - `LAB_TESTING_ROOT`: Path to lab testing framework (default: `/data_drive/esl/ai-lab-testing`)
 - `VPN_CONFIG_PATH`: Path to WireGuard config file (optional, auto-detected if not set)
+- `TARGET_NETWORK`: Target network for lab testing operations (default: `192.168.2.0/24`)
+- `MCP_DEV_MODE`: Enable development mode with auto-reload (set to `1`, `true`, or `yes` to enable)
+
+### Target Network Configuration
+
+The target network determines which network is used for lab testing operations. It can be configured in three ways (priority order):
+
+1. **Environment Variable**: Set `TARGET_NETWORK` environment variable
+   ```bash
+   export TARGET_NETWORK=192.168.2.0/24
+   ```
+
+2. **Config File**: Add to `lab_devices.json`:
+   ```json
+   {
+     "lab_infrastructure": {
+       "network_access": {
+         "target_network": "192.168.2.0/24",
+         "lab_networks": ["192.168.2.0/24"]
+       }
+     }
+   }
+   ```
+
+3. **Default**: `192.168.2.0/24` (if not configured)
+
+The `lab_networks` list can contain multiple networks for scanning, but `target_network` specifies the primary network for lab operations.
 
 ### VPN Configuration
 
@@ -51,7 +78,10 @@ Add to `~/.cursor/mcp.json`:
     "ai-lab-testing": {
       "command": "python3.10",
       "args": ["/absolute/path/to/lab_testing/server.py"],
-      "env": {"LAB_TESTING_ROOT": "/data_drive/esl/ai-lab-testing"}
+      "env": {
+        "LAB_TESTING_ROOT": "/data_drive/esl/ai-lab-testing",
+        "MCP_DEV_MODE": "1"
+      }
     }
   }
 }
@@ -61,7 +91,30 @@ Add to `~/.cursor/mcp.json`:
 
 Or use installed package: `"command": "python3.10", "args": ["-m", "lab_testing.server"]`
 
-Restart Cursor.
+### Development Mode (Auto-Reload)
+
+During development, you can enable auto-reload so that code changes are picked up automatically without restarting Cursor. This is especially useful when modifying tool handlers or network mapping code.
+
+**To enable:** Add `"MCP_DEV_MODE": "1"` to the `env` section in your MCP configuration (as shown above).
+
+**How it works:**
+- The server checks for file changes before each tool call
+- Modified Python modules are automatically reloaded using `importlib.reload()`
+- No need to restart Cursor after making code changes
+- Reloaded modules are logged for debugging
+
+**Note:** Auto-reload only works for modules in the `lab_testing` package. Changes to `server.py` itself still require a restart.
+
+Restart Cursor after initial setup.
+
+### Tool Call Timeouts
+
+**Note:** Some tools (like `create_network_map`) may take longer than 30 seconds if they perform network scans. Cursor has a default 30-second timeout for MCP tool calls.
+
+**Solutions:**
+1. **Use Quick Mode**: For `create_network_map`, set `quick_mode: true` to skip network scanning and only show configured devices (completes in <5 seconds).
+2. **Optimized Scanning**: The network scanner has been optimized with faster ping timeouts (0.5s) and increased parallelism (100 workers) to reduce scan time.
+3. **Client Timeout**: Currently, there's no way to configure the MCP client timeout in Cursor's `mcp.json`. The timeout is hardcoded in the Cursor client. If you need full network scans, consider running the tool directly via Python or use quick mode for faster results.
 
 ## Verification
 

lab_testing/config.py

@@ -27,6 +27,10 @@
 # If not set, searches common locations
 VPN_CONFIG_PATH_ENV = os.getenv("VPN_CONFIG_PATH")
 
+# Target network configuration - can be overridden via TARGET_NETWORK environment variable
+# Default target network for lab testing operations
+DEFAULT_TARGET_NETWORK = os.getenv("TARGET_NETWORK", "192.168.2.0/24")
+
 
 def get_lab_devices_config() -> Path:
     """Get path to lab devices configuration file"""
@@ -99,6 +103,72 @@ def get_logs_dir() -> Path:
     return LOGS_DIR
 
 
+def get_target_network() -> str:
+    """
+    Get the target network for lab testing operations.
+    
+    Priority:
+    1. TARGET_NETWORK environment variable
+    2. Value from lab_devices.json config file (if exists)
+    3. Default: 192.168.2.0/24
+    
+    Returns:
+        Network CIDR string (e.g., "192.168.2.0/24")
+    """
+    # Check environment variable first
+    env_network = os.getenv("TARGET_NETWORK")
+    if env_network:
+        return env_network
+    
+    # Check config file
+    try:
+        if LAB_DEVICES_JSON.exists():
+            import json
+            with open(LAB_DEVICES_JSON) as f:
+                config = json.load(f)
+                infrastructure = config.get("lab_infrastructure", {})
+                network_access = infrastructure.get("network_access", {})
+                target_network = network_access.get("target_network")
+                if target_network:
+                    return target_network
+    except Exception:
+        # If config read fails, fall back to default
+        pass
+    
+    # Default target network
+    return DEFAULT_TARGET_NETWORK
+
+
+def get_lab_networks() -> list[str]:
+    """
+    Get list of lab networks for scanning.
+    
+    Priority:
+    1. Networks from lab_devices.json config file
+    2. TARGET_NETWORK environment variable (as single-item list)
+    3. Default: [target_network]
+    
+    Returns:
+        List of network CIDR strings
+    """
+    # Check config file first
+    try:
+        if LAB_DEVICES_JSON.exists():
+            import json
+            with open(LAB_DEVICES_JSON) as f:
+                config = json.load(f)
+                infrastructure = config.get("lab_infrastructure", {})
+                network_access = infrastructure.get("network_access", {})
+                lab_networks = network_access.get("lab_networks")
+                if lab_networks and isinstance(lab_networks, list):
+                    return lab_networks
+    except Exception:
+        pass
+    
+    # Fall back to target network
+    return [get_target_network()]
+
+
 def validate_config() -> tuple:
     """Validate that required configuration files exist"""
     errors = []

lab_testing/server.py

@@ -15,7 +15,7 @@
 import sys
 import time
 import uuid
-from typing import Any, Dict, List
+from typing import Any, Dict, List, Union
 
 # MCP SDK imports
 # Note: MCP SDK structure may vary - adjust imports based on actual SDK version
@@ -59,6 +59,13 @@
 setup_logger()
 logger = get_logger()
 
+# Development mode: Set up auto-reload
+try:
+    from lab_testing.server.dev_reload import setup_auto_reload
+    setup_auto_reload()
+except ImportError:
+    pass  # dev_reload module not available
+
 # Initialize MCP server
 server = Server("ai-lab-testing-mcp")
 
@@ -82,7 +89,7 @@ async def handle_list_tools() -> List[Tool]:
 
 
 @server.call_tool()
-async def handle_call_tool(name: str, arguments: Dict[str, Any]) -> List[TextContent]:
+async def handle_call_tool(name: str, arguments: Dict[str, Any]) -> List[Union[TextContent, ImageContent]]:
     """Handle tool execution requests"""
     request_id = str(uuid.uuid4())[:8]
     start_time = time.time()