@nclamvn/vibecode-cli 1.3.0 → 1.6.0

package/src/commands/build.js

@@ -22,6 +22,8 @@ import { STATES } from '../config/constants.js';
 import { getBuildReportTemplate } from '../config/templates.js';
 import { ensureDir, pathExists, appendToFile, readMarkdown, writeJson } from '../utils/files.js';
 import { printBox, printError, printSuccess, printWarning, printNextStep } from '../ui/output.js';
+import { showError, inlineError } from '../ui/error-translator.js';
+import { BackupManager } from '../core/backup.js';
 import {
   spawnClaudeCode,
   isClaudeCodeAvailable,
@@ -75,7 +77,7 @@ export async function buildCommand(options = {}) {
     }
 
   } catch (error) {
-    printError(error.message);
+    showError(error, { verbose: options.verbose });
     process.exit(1);
   }
 }
@@ -85,6 +87,10 @@ export async function buildCommand(options = {}) {
  * "Contract LOCKED = License to build"
  */
 async function handleAutoBuild(currentState, projectName, sessionId, sessionPath, specHash, options) {
+  // Create backup before build
+  const backup = new BackupManager();
+  await backup.createBackup('build-auto');
+
   // Check state - must be PLAN_CREATED or BUILD_IN_PROGRESS or REVIEW_FAILED
   const validStates = [STATES.PLAN_CREATED, STATES.BUILD_IN_PROGRESS, STATES.REVIEW_FAILED];
   if (!validStates.includes(currentState)) {
@@ -227,7 +233,7 @@ ${evidence.screenshots > 0 ? `  ✅ ${evidence.screenshots} screenshots` : '
 
   } catch (error) {
     await appendToFile(logPath, `\nERROR: ${error.message}\n`);
-    printError(`Auto build failed: ${error.message}`);
+    showError(error);
     process.exit(1);
   }
 }
@@ -240,6 +246,10 @@ async function handleIterativeBuild(currentState, projectName, sessionId, sessio
   const maxIterations = options.max || 3;
   const strictMode = options.strict || false;
 
+  // Create backup before iterative build
+  const backup = new BackupManager();
+  await backup.createBackup('build-iterate');
+
   // Check state - must be PLAN_CREATED or BUILD_IN_PROGRESS or REVIEW_FAILED
   const validStates = [STATES.PLAN_CREATED, STATES.BUILD_IN_PROGRESS, STATES.REVIEW_FAILED];
   if (!validStates.includes(currentState)) {
@@ -451,7 +461,7 @@ Starting build-test-fix loop...`;
       }
 
     } catch (error) {
-      console.log(chalk.red(`\n❌ Build error: ${error.message}`));
+      console.log('\n' + inlineError(error));
       await logIteration(logPath, iteration, `Error: ${error.message}`);
 
       iterationState = finalizeIterationState(iterationState, 'error');

package/src/commands/debug.js

@@ -0,0 +1,457 @@
+// ═══════════════════════════════════════════════════════════════════════════════
+// VIBECODE CLI - Debug Command
+// Phase G: Intelligent 9-Step Bug Fixing
+// ═══════════════════════════════════════════════════════════════════════════════
+
+import chalk from 'chalk';
+import ora from 'ora';
+import path from 'path';
+import fs from 'fs-extra';
+import readline from 'readline';
+import { printBox, printError, printSuccess, printWarning, printNextStep } from '../ui/output.js';
+import { createDebugEngine } from '../debug/index.js';
+
+/**
+ * Debug Command Entry Point
+ * Usage:
+ *   vibecode debug                         - Interactive mode (default)
+ *   vibecode debug "error description"     - Debug with description
+ *   vibecode debug --log                   - Paste error log
+ *   vibecode debug --image <path>          - Debug from screenshot
+ *   vibecode debug --auto                  - Auto-scan and fix
+ */
+export async function debugCommand(args = [], options = {}) {
+  try {
+    const projectPath = process.cwd();
+
+    // Parse command arguments
+    const input = parseDebugInput(args, options);
+
+    // If interactive mode or no input provided, enter interactive mode
+    if (options.interactive || (!input.description && !input.log && !input.image && !input.auto)) {
+      return interactiveDebug(projectPath, options);
+    }
+
+    // Create debug engine
+    const engine = createDebugEngine(projectPath, {
+      autoFix: options.autoFix !== false,
+      maxAttempts: options.attempts || 3,
+      verbose: options.verbose || false,
+      interactive: !options.quiet
+    });
+
+    // Show intro
+    console.log();
+    printBox(`🔍 VIBECODE DEBUG
+
+Intelligent 9-Step Bug Fixing
+Mode: ${input.auto ? 'Auto-Scan' : input.log ? 'Log Analysis' : input.image ? 'Image Analysis' : 'Description'}
+Max Attempts: ${options.attempts || 3}`, { borderColor: 'cyan' });
+    console.log();
+
+    // Run debug workflow
+    const result = await engine.debug(input);
+
+    // Show final result
+    await showDebugResult(result, projectPath);
+
+  } catch (error) {
+    printError(`Debug failed: ${error.message}`);
+    if (options.verbose) {
+      console.error(error.stack);
+    }
+    process.exit(1);
+  }
+}
+
+/**
+ * Parse debug input from args and options
+ */
+function parseDebugInput(args, options) {
+  const input = {
+    description: null,
+    log: null,
+    image: null,
+    auto: options.auto || false
+  };
+
+  // Join args as description
+  if (args.length > 0) {
+    input.description = args.join(' ');
+  }
+
+  // Log paste mode
+  if (options.log) {
+    if (typeof options.log === 'string') {
+      input.log = options.log;
+    } else {
+      // Will need to read from stdin or file
+      input.log = options.logContent || null;
+    }
+  }
+
+  // Image mode
+  if (options.image) {
+    input.image = options.image;
+  }
+
+  return input;
+}
+
+/**
+ * Show debug result summary
+ */
+async function showDebugResult(result, projectPath) {
+  console.log();
+
+  if (result.status === 'resolved') {
+    const content = `✅ BUG FIXED!
+
+Status: ${result.status}
+Attempts: ${result.session.attempts}
+Steps: ${result.session.steps}
+
+Documentation saved to:
+  .vibecode/debug/fixes.md
+
+Prevention rules added to:
+  CLAUDE.md`;
+
+    printBox(content, { borderColor: 'green' });
+    printNextStep('Build your project to verify the fix');
+
+  } else if (result.status === 'no_error') {
+    printSuccess('No errors found to fix!');
+
+  } else if (result.status === 'no_hypothesis') {
+    printWarning('Could not determine how to fix this error.');
+    console.log(chalk.gray('\nTry:'));
+    console.log(chalk.gray('  • Providing more details about the error'));
+    console.log(chalk.gray('  • Pasting the full error log with --log'));
+    console.log(chalk.gray('  • Running vibecode debug --auto to scan project'));
+
+  } else {
+    // Show escalation options - user NEVER bế tắc
+    console.log(chalk.yellow(`
+╭───────────────────────────────────────────────────────────────────╮
+│                                                                   │
+│   ⚠️  COULD NOT AUTO-RESOLVE                                      │
+│                                                                   │
+│   Status: ${result.status.padEnd(47)}│
+│   Attempts: ${String(result.session.attempts).padEnd(45)}│
+│                                                                   │
+│   Don't worry! You have options:                                  │
+│                                                                   │
+│   1. vibecode assist "describe your issue"                       │
+│      → Direct AI expert access with full context                 │
+│                                                                   │
+│   2. vibecode debug -i                                           │
+│      → Interactive debug with more context                       │
+│                                                                   │
+│   3. vibecode debug --attempts 5                                 │
+│      → Retry with more attempts                                  │
+│                                                                   │
+╰───────────────────────────────────────────────────────────────────╯
+`));
+
+    // Ask if user wants to auto-escalate to summon
+    const rl = readline.createInterface({
+      input: process.stdin,
+      output: process.stdout
+    });
+
+    try {
+      const answer = await new Promise(resolve => {
+        rl.question(chalk.cyan('\n🤝 Escalate to AI Assist? (Y/n): '), resolve);
+      });
+      rl.close();
+
+      if (answer.toLowerCase() !== 'n') {
+        const { assistCommand } = await import('./assist.js');
+        const errorMsg = result.message || 'the error that occurred';
+        await assistCommand([`Fix ${errorMsg}`], {});
+      }
+    } catch {
+      rl.close();
+    }
+  }
+}
+
+/**
+ * Show debug command help
+ */
+function showDebugHelp() {
+  console.log();
+  printBox(`🔍 VIBECODE DEBUG - Help
+
+Intelligent 9-step bug fixing powered by AI.
+
+Usage:
+  vibecode debug "error description"     Quick fix from description
+  vibecode debug --auto                  Auto-scan project for errors
+  vibecode debug --log "error text"      Fix from error log
+  vibecode debug --image path/to/img     Fix from screenshot
+
+Options:
+  --auto             Auto-scan mode - find and fix errors
+  --log <text>       Provide error log directly
+  --image <path>     Provide error screenshot
+  --attempts <n>     Max fix attempts (default: 3)
+  --verbose          Show detailed output
+  --quiet            Minimal output
+
+Examples:
+  vibecode debug "Cannot read property of undefined"
+  vibecode debug --auto
+  vibecode debug --log "TypeError: x is not a function"
+  vibecode debug --attempts 5 "module not found"`, { borderColor: 'cyan' });
+  console.log();
+}
+
+/**
+ * Interactive Debug Mode
+ * REPL-like interface for debugging
+ */
+async function interactiveDebug(projectPath, options) {
+  const engine = createDebugEngine(projectPath, {
+    autoFix: true,
+    maxAttempts: options.attempts || 3,
+    verbose: options.verbose || false,
+    interactive: true
+  });
+
+  // Show welcome banner
+  console.log(chalk.cyan(`
+╭────────────────────────────────────────────────────────────────────╮
+│                                                                    │
+│   🔍 VIBECODE DEBUG (Interactive Mode)                             │
+│                                                                    │
+│   Commands:                                                        │
+│   • Type error description or paste log                            │
+│   • /scan   - Auto-scan project for errors                         │
+│   • /fix    - Apply last suggested fix                             │
+│   • /retry  - Retry with current context                           │
+│   • /clear  - Clear context                                        │
+│   • /help   - Show help                                            │
+│   • /quit   - Exit                                                 │
+│                                                                    │
+╰────────────────────────────────────────────────────────────────────╯
+  `));
+
+  const rl = readline.createInterface({
+    input: process.stdin,
+    output: process.stdout,
+    prompt: chalk.green('debug> ')
+  });
+
+  // Session context
+  const context = {
+    errors: [],
+    lastAnalysis: null,
+    lastFix: null,
+    lastEvidence: null
+  };
+
+  rl.prompt();
+
+  rl.on('line', async (line) => {
+    const input = line.trim();
+
+    if (!input) {
+      rl.prompt();
+      return;
+    }
+
+    // Handle commands
+    if (input.startsWith('/')) {
+      await handleDebugCommand(input, engine, context, rl);
+    } else {
+      // Treat as error description or log
+      await handleDebugInput(input, engine, context);
+    }
+
+    rl.prompt();
+  });
+
+  rl.on('close', () => {
+    console.log(chalk.cyan('\n👋 Debug session ended.\n'));
+    process.exit(0);
+  });
+}
+
+/**
+ * Handle debug commands (/scan, /fix, etc.)
+ */
+async function handleDebugCommand(cmd, engine, context, rl) {
+  const command = cmd.toLowerCase().split(' ')[0];
+
+  switch (command) {
+    case '/scan':
+      console.log(chalk.blue('\n🔍 Scanning project for errors...\n'));
+      try {
+        const result = await engine.debug({ auto: true });
+        context.lastAnalysis = result;
+        context.lastEvidence = result.evidence;
+
+        if (result.status === 'resolved') {
+          console.log(chalk.green('\n✅ All errors resolved!'));
+        } else if (result.status === 'no_error') {
+          console.log(chalk.green('\n✅ No errors found!'));
+        } else {
+          console.log(chalk.yellow('\n⚠️  Some issues found. Use /fix to apply fixes.'));
+        }
+      } catch (error) {
+        console.log(chalk.red(`\nScan error: ${error.message}`));
+      }
+      break;
+
+    case '/fix':
+      if (context.lastAnalysis && context.lastAnalysis.fixPrompt) {
+        console.log(chalk.blue('\n🔧 Applying fix...\n'));
+        try {
+          const fixResult = await engine.applyFix(context.lastAnalysis.fixPrompt);
+          if (fixResult.success) {
+            console.log(chalk.green('\n✅ Fix applied! Run /scan to verify.'));
+          } else {
+            console.log(chalk.yellow(`\n⚠️  Fix may have issues: ${fixResult.error || 'Check output above'}`));
+          }
+        } catch (error) {
+          console.log(chalk.red(`\nFix error: ${error.message}`));
+        }
+      } else if (context.errors.length > 0) {
+        // Try to generate fix from accumulated context
+        console.log(chalk.blue('\n🔧 Generating fix from context...\n'));
+        try {
+          const result = await engine.debug({ description: context.errors.join('\n\n') });
+          context.lastAnalysis = result;
+        } catch (error) {
+          console.log(chalk.red(`\nError: ${error.message}`));
+        }
+      } else {
+        console.log(chalk.yellow('\nNo analysis to fix. Run /scan or describe an error first.'));
+      }
+      break;
+
+    case '/retry':
+      if (context.errors.length > 0) {
+        console.log(chalk.blue('\n🔄 Retrying with accumulated context...\n'));
+        try {
+          const result = await engine.debug({ description: context.errors.join('\n\n') });
+          context.lastAnalysis = result;
+        } catch (error) {
+          console.log(chalk.red(`\nRetry error: ${error.message}`));
+        }
+      } else {
+        console.log(chalk.yellow('\nNo context to retry with. Describe an error first.'));
+      }
+      break;
+
+    case '/clear':
+      context.errors = [];
+      context.lastAnalysis = null;
+      context.lastFix = null;
+      context.lastEvidence = null;
+      console.log(chalk.green('\n✓ Context cleared.\n'));
+      break;
+
+    case '/context':
+      console.log(chalk.cyan('\n📋 Current Context:'));
+      console.log(chalk.gray(`   Errors collected: ${context.errors.length}`));
+      console.log(chalk.gray(`   Last analysis: ${context.lastAnalysis ? 'Yes' : 'No'}`));
+      if (context.errors.length > 0) {
+        console.log(chalk.gray('\n   Recent errors:'));
+        context.errors.slice(-3).forEach((e, i) => {
+          console.log(chalk.gray(`   ${i + 1}. ${e.substring(0, 60)}${e.length > 60 ? '...' : ''}`));
+        });
+      }
+      console.log();
+      break;
+
+    case '/help':
+      console.log(chalk.cyan(`
+  Commands:
+    /scan     Auto-scan project for errors
+    /fix      Apply last suggested fix
+    /retry    Retry with accumulated context
+    /clear    Clear all context
+    /context  Show current context
+    /help     Show this help
+    /quit     Exit debug mode
+
+  Tips:
+    • Paste error messages directly for analysis
+    • Describe what's wrong in plain English
+    • Use /scan for automatic error detection
+    • Context accumulates - use /clear to reset
+      `));
+      break;
+
+    case '/quit':
+    case '/exit':
+    case '/q':
+      rl.close();
+      break;
+
+    default:
+      console.log(chalk.yellow(`\nUnknown command: ${cmd}`));
+      console.log(chalk.gray('Type /help for available commands.\n'));
+  }
+}
+
+/**
+ * Handle user input (error description or log)
+ */
+async function handleDebugInput(input, engine, context) {
+  // Add to context
+  context.errors.push(input);
+
+  // Show spinner
+  const spinner = ora('Analyzing error...').start();
+
+  try {
+    const result = await engine.debug({ description: input });
+    context.lastAnalysis = result;
+
+    spinner.stop();
+
+    if (result.status === 'resolved') {
+      console.log(chalk.green('\n✅ Error analyzed and fixed!\n'));
+    } else if (result.status === 'no_error') {
+      console.log(chalk.green('\n✅ No actionable error found.\n'));
+    } else {
+      console.log(chalk.yellow('\n⚠️  Analysis complete. Use /fix to apply suggested fix.\n'));
+
+      // Show brief analysis
+      if (result.session && result.session.steps > 0) {
+        console.log(chalk.gray(`   Steps completed: ${result.session.steps}`));
+        console.log(chalk.gray(`   Attempts: ${result.session.attempts}`));
+      }
+    }
+  } catch (error) {
+    spinner.stop();
+    console.log(chalk.red(`\nAnalysis error: ${error.message}\n`));
+  }
+}
+
+/**
+ * Interactive log paste mode (for future enhancement)
+ */
+async function interactiveLogMode() {
+  console.log(chalk.cyan('Paste your error log below (end with Ctrl+D):'));
+  console.log(chalk.gray('---'));
+
+  let logContent = '';
+
+  process.stdin.setEncoding('utf8');
+
+  return new Promise((resolve) => {
+    process.stdin.on('data', (chunk) => {
+      logContent += chunk;
+    });
+
+    process.stdin.on('end', () => {
+      console.log(chalk.gray('---'));
+      resolve(logContent.trim());
+    });
+  });
+}