SSH MCP Server

/** * Terminal Display Diagnosis using Villenele Framework * * This test uses the comprehensive Villenele Terminal History Testing Framework * to diagnose the exact root cause of terminal display issues: * 1. Commands appearing on separate lines instead of inline with prompts * 2. Corrupted prompts (missing characters like '[j' -> 'sbattig@localhost') * 3. Terminal control sequence issues causing positioning problems * * CRITICAL: Uses real WebSocket capture from browser perspective - NO MOCKS */ import { JestTestUtilities, WebSocketValidationResult } from './integration/terminal-history-framework/jest-test-utilities'; import { CommandConfigurationJSON } from './integration/terminal-history-framework/flexible-command-configuration'; import { WorkflowResult } from './integration/terminal-history-framework/comprehensive-response-collector'; describe('Terminal Display Diagnosis with Villenele Framework', () => { let testUtils: JestTestUtilities; beforeAll(async () => { // Enable detailed logging to see exactly what's happening testUtils = new JestTestUtilities({ enableDetailedLogging: true, enableErrorDiagnostics: true, testTimeout: 45000 }); // Extend Jest matchers for terminal validation JestTestUtilities.extendJestMatchers(); }); beforeEach(async () => { await testUtils.setupTest('terminal-display-diagnosis'); }); afterEach(async () => { await testUtils.cleanupTest(); }); describe('Root Cause Analysis - WebSocket Message Capture', () => { test('should capture exact WebSocket messages from complete-fix-test session to diagnose prompt issues', async () => { console.log('\n=== TERMINAL DISPLAY DIAGNOSIS STARTING ==='); console.log('Testing session: complete-fix-test'); console.log('Expected URL: http://localhost:8084/session/complete-fix-test'); console.log('Problem: Commands on separate lines + corrupted prompts'); const config: CommandConfigurationJSON = { preWebSocketCommands: [ // Connect to SSH session 'ssh_connect {"name": "complete-fix-test", "host": "localhost", "username": "jsbattig", "keyFilePath": "~/.ssh/id_ed25519"}', // Execute the exact commands that are showing display issues 'ssh_exec {"sessionName": "complete-fix-test", "command": "echo \\"final validation test\\""}', // Execute a few more commands to see the pattern 'ssh_exec {"sessionName": "complete-fix-test", "command": "pwd"}', 'ssh_exec {"sessionName": "complete-fix-test", "command": "whoami"}' ], postWebSocketCommands: [ // Test real-time command execution after WebSocket connection {initiator: 'mcp-client', command: 'ssh_exec {"sessionName": "complete-fix-test", "command": "date"}'}, {initiator: 'mcp-client', command: 'ssh_exec {"sessionName": "complete-fix-test", "command": "hostname"}'} ], workflowTimeout: 40000, sessionName: 'complete-fix-test' }; let workflowResult: WorkflowResult; try { console.log('\n=== EXECUTING VILLENELE FRAMEWORK ==='); workflowResult = await testUtils.runTerminalHistoryTest(config); console.log('\n=== WORKFLOW EXECUTION RESULTS ==='); console.log(`Success: ${workflowResult.success}`); console.log(`Total Execution Time: ${workflowResult.totalExecutionTime}ms`); console.log(`Phase Breakdown Available: ${workflowResult.phaseBreakdown ? 'Yes' : 'No'}`); if (workflowResult.phaseBreakdown) { console.log(` - Server Launch: ${workflowResult.phaseBreakdown.serverLaunchSuccess}`); console.log(` - Pre-WebSocket Commands: ${workflowResult.phaseBreakdown.preWebSocketCommandsSuccess}`); console.log(` - WebSocket Connection: ${workflowResult.phaseBreakdown.webSocketConnectionSuccess}`); console.log(` - History Replay Capture: ${workflowResult.phaseBreakdown.historyReplayCaptureSuccess}`); console.log(` - Post-WebSocket Commands: ${workflowResult.phaseBreakdown.postWebSocketCommandsSuccess}`); } // This is the CRITICAL data - the verbatim WebSocket messages console.log(`\n=== RAW WEBSOCKET MESSAGES (VERBATIM) ===`); console.log(`Message Length: ${workflowResult.concatenatedResponses?.length || 0} characters`); if (workflowResult.concatenatedResponses) { // Show raw messages with visible control characters const rawMessages = workflowResult.concatenatedResponses; console.log('Raw messages with escaped control chars:'); console.log(JSON.stringify(rawMessages, null, 2)); // Show messages line by line for analysis console.log('\n=== LINE-BY-LINE ANALYSIS ==='); const lines = rawMessages.split('\n'); lines.forEach((line, index) => { console.log(`Line ${index + 1}: "${line}"`); console.log(` - Contains \\r: ${line.includes('\r')}`); console.log(` - Length: ${line.length}`); console.log(` - Ends with $: ${line.endsWith('$')}`); console.log(` - Contains prompt pattern: ${/\[[a-zA-Z0-9_-]+@[a-zA-Z0-9_-]+\s+[^\]]+\]\$/.test(line)}`); }); console.log('\n=== PROMPT PATTERN ANALYSIS ==='); const promptMatches = rawMessages.match(/\[[a-zA-Z0-9_-]+@[a-zA-Z0-9_-]+\s+[^\]]+\]\$/g); console.log(`Found prompts: ${promptMatches?.length || 0}`); if (promptMatches) { promptMatches.forEach((prompt, index) => { console.log(`Prompt ${index + 1}: "${prompt}"`); }); } console.log('\n=== COMMAND DETECTION ANALYSIS ==='); const commandPattern = /echo "final validation test"|pwd|whoami|date|hostname/g; const commandMatches = rawMessages.match(commandPattern); console.log(`Found commands: ${commandMatches?.length || 0}`); if (commandMatches) { commandMatches.forEach((cmd, index) => { console.log(`Command ${index + 1}: "${cmd}"`); }); } console.log('\n=== POSITIONING ISSUE ANALYSIS ==='); // Look for the specific issues mentioned by the user // Check for commands on separate lines (should be inline with prompts) const separateLinePattern = /\]\$\s*\n\s*[a-zA-Z]/g; const separateLineMatches = rawMessages.match(separateLinePattern); console.log(`Commands on separate lines: ${separateLineMatches?.length || 0}`); if (separateLineMatches) { separateLineMatches.forEach((match, index) => { console.log(`Separate line issue ${index + 1}: "${match}"`); }); } // Check for corrupted prompts (missing characters) const corruptedPromptPattern = /sbattig@localhost/g; const corruptedMatches = rawMessages.match(corruptedPromptPattern); console.log(`Corrupted prompts (missing [j): ${corruptedMatches?.length || 0}`); if (corruptedMatches) { console.log('CRITICAL: Found corrupted prompts - missing opening bracket and username prefix!'); } console.log('\n=== CRLF ANALYSIS ==='); console.log(`Contains CRLF (\\r\\n): ${rawMessages.includes('\r\n')}`); console.log(`Contains LF only (\\n): ${rawMessages.includes('\n')}`); console.log(`CRLF count: ${(rawMessages.match(/\r\n/g) || []).length}`); console.log(`LF count: ${(rawMessages.match(/\n/g) || []).length}`); } if (workflowResult.error) { console.log(`\n=== ERROR INFORMATION ===`); console.log(`Error: ${workflowResult.error}`); } } catch (error) { console.error('\n=== VILLENELE FRAMEWORK ERROR ==='); console.error('Error during diagnosis:', error); throw error; } // Validate the workflow succeeded expect(workflowResult.success).toBe(true); expect(workflowResult.concatenatedResponses).toBeDefined(); expect(workflowResult.concatenatedResponses!.length).toBeGreaterThan(0); // Use Villenele's WebSocket validation const validation: WebSocketValidationResult = testUtils.validateWebSocketMessages(workflowResult.concatenatedResponses!); console.log('\n=== VILLENELE VALIDATION RESULTS ==='); console.log(`Has Messages: ${validation.hasMessages}`); console.log(`Has CRLF: ${validation.hasCRLF}`); console.log(`Has Prompts: ${validation.hasPrompts}`); console.log(`Command Count: ${validation.commandCount}`); console.log(`Message Count: ${validation.messageCount}`); console.log(`Validation Errors: ${validation.errors.length}`); if (validation.errors.length > 0) { console.log('=== VALIDATION ERRORS ==='); validation.errors.forEach((error, index) => { console.log(`Error ${index + 1}: ${error}`); }); } // Store the results for further analysis console.log('\n=== DIAGNOSIS COMPLETE ==='); console.log('WebSocket messages captured successfully.'); console.log('Raw data available for root cause analysis.'); console.log('Check the console output above for detailed terminal sequence analysis.'); }); test('should analyze specific ANSI control sequences causing positioning issues', async () => { console.log('\n=== ANSI CONTROL SEQUENCE ANALYSIS ==='); const config: CommandConfigurationJSON = { preWebSocketCommands: [ 'ssh_connect {"name": "ansi-analysis", "host": "localhost", "username": "jsbattig", "keyFilePath": "~/.ssh/id_ed25519"}', // Simple command to analyze exact output 'ssh_exec {"sessionName": "ansi-analysis", "command": "echo \\"test command\\""}', ], postWebSocketCommands: [ {initiator: 'mcp-client', command: 'ssh_exec {"sessionName": "ansi-analysis", "command": "pwd"}'} ], workflowTimeout: 30000, sessionName: 'ansi-analysis' }; const workflowResult = await testUtils.runTerminalHistoryTest(config); expect(workflowResult.success).toBe(true); if (workflowResult.concatenatedResponses) { console.log('\n=== ANSI ESCAPE SEQUENCE DETECTION ==='); const rawMessages = workflowResult.concatenatedResponses; // Look for common ANSI escape sequences const ansiPatterns = { 'Cursor Movement': new RegExp('\u001b\[[0-9;]*[HfABCD]', 'g'), 'Clear Screen/Line': new RegExp('\u001b\[[0-9;]*[JK]', 'g'), 'Color Codes': new RegExp('\u001b\[[0-9;]*m', 'g'), 'Save/Restore Cursor': new RegExp('\u001b\[[su]', 'g'), 'Set Mode': new RegExp('\u001b\[[?][0-9;]*[hl]', 'g') }; Object.entries(ansiPatterns).forEach(([name, pattern]) => { const matches = rawMessages.match(pattern); console.log(`${name}: ${matches?.length || 0} sequences found`); if (matches && matches.length > 0) { console.log(` Examples: ${matches.slice(0, 3).map(m => JSON.stringify(m)).join(', ')}`); } }); // Check for carriage returns without newlines (could cause overwriting) const crWithoutLf = rawMessages.match(/\r(?!\n)/g); console.log(`Carriage Returns without LF: ${crWithoutLf?.length || 0}`); if (crWithoutLf && crWithoutLf.length > 0) { console.log('CRITICAL: Found CR without LF - this can cause character overwriting!'); } } }); }); describe('Fix Implementation Phase', () => { test('should apply terminal control sequence fixes and validate results', async () => { // This test will be implemented after root cause analysis identifies the issues console.log('\n=== FIX IMPLEMENTATION PHASE ==='); console.log('This test will implement fixes based on Villenele diagnosis results'); console.log('Currently pending root cause identification...'); // Placeholder for fix implementation expect(true).toBe(true); }); }); });