SSH MCP Server

# Feature: Nuclear Fallback ## Feature Description Emergency recovery system that kills and re-establishes SSH sessions when command cancellation fails within 30 seconds, ensuring the system never hangs permanently regardless of command behavior. ## Technical Architecture ### Fallback Trigger - 30-second timer starts when cancellation command issued - Applies to both browser and MCP cancellation requests - Automatic SSH session termination and transparent reconnection ### Recovery Process ``` Cancellation Timeout → SSH Connection Kill → Session Re-establishment → State Cleanup → Client Notification ``` ## Story Implementation Order - [ ] 01_Story_CancellationTimeoutDetection - [ ] 02_Story_SSHSessionTermination - [ ] 03_Story_TransparentReconnection - [ ] 04_Story_StateCleanupAndNotification ## Dependencies **Prerequisites**: - 01_Feat_BrowserCommandTracking (for state cleanup) - 02_Feat_MCPCommandGating (for gatekeeper reset) - 03_Feat_CommandCancellation (triggers fallback mechanism) ## Success Criteria - 🚀 **Never Hangs**: System recovers from any hung command within 30 seconds - ✅ **Transparent Recovery**: Users/clients experience seamless reconnection - 📊 **Complete Cleanup**: All command state cleared after nuclear reset - ⚙️ **Gatekeeper Reset**: Command gating restored after session recovery This feature ensures system reliability by providing guaranteed recovery from any command that becomes unresponsive to normal cancellation mechanisms.