claude-flow-novice 1.5.2 → 1.5.3

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "claude-flow-novice",
-  "version": "1.5.2",
+  "version": "1.5.3",
   "description": "Standalone Claude Flow for beginners - AI agent orchestration made easy with enhanced TDD testing pipeline. Enhanced init command creates complete agent system, MCP configuration with 30 essential tools, and automated hooks with single-file testing, real-time coverage analysis, and advanced validation. Fully standalone with zero external dependencies, complete project setup in one command.",
   "mcpName": "io.github.ruvnet/claude-flow",
   "main": ".claude-flow-novice/dist/index.js",
@@ -39,9 +39,9 @@
     "optimize:validate:hardware": "node scripts/optimization/config-validator.js validate hardware",
     "optimize:validate:monitoring": "node scripts/optimization/config-validator.js validate monitoring",
     "build": "scripts/build/unified-builder.sh safe",
-    "build:swc": "swc src -d .claude-flow-novice/dist --only='**/*.ts' --config-file .swcrc && cp -r src/slash-commands .claude-flow-novice/dist/src/ && cp -r src/cli/simple-commands/hooks .claude-flow-novice/dist/src/cli/simple-commands/ && cp -r .claude/agents .claude-flow-novice/.claude/",
+    "build:swc": "swc src -d .claude-flow-novice/dist --only='**/*.ts' --config-file .swcrc && cp -r src/slash-commands .claude-flow-novice/dist/src/ && cp -r src/cli/simple-commands/hooks .claude-flow-novice/dist/src/cli/simple-commands/ && cp -r src/cli/simple-commands/init/templates .claude-flow-novice/dist/src/cli/simple-commands/init/ && cp -r .claude/agents .claude-flow-novice/.claude/",
     "build:types": "tsc --project config/typescript/tsconfig.json --emitDeclarationOnly --outDir .claude-flow-novice/dist --skipLibCheck",
-    "build:watch": "swc src -d .claude-flow-novice/dist --watch --config-file .swcrc && cp -r src/slash-commands .claude-flow-novice/dist/src/ && cp -r src/cli/simple-commands/hooks .claude-flow-novice/dist/src/cli/simple-commands/ && npm run copy:agents",
+    "build:watch": "swc src -d .claude-flow-novice/dist --watch --config-file .swcrc && cp -r src/slash-commands .claude-flow-novice/dist/src/ && cp -r src/cli/simple-commands/hooks .claude-flow-novice/dist/src/cli/simple-commands/ && cp -r src/cli/simple-commands/init/templates .claude-flow-novice/dist/src/cli/simple-commands/init/ && npm run copy:agents",
     "build:legacy": "scripts/build/unified-builder.sh migration",
     "build:workaround": "scripts/build/unified-builder.sh workaround",
     "build:force": "scripts/build/unified-builder.sh force",

package/src/cli/simple-commands/init/index.js

@@ -1,6 +1,9 @@
 // init/index.js - Initialize Claude Code integration files
 import { printSuccess, printError, printWarning, exit } from '../../utils.js';
 import { existsSync } from 'fs';
+import { promises as fs } from 'fs';
+import { fileURLToPath } from 'url';
+import { dirname, join } from 'path';
 import process from 'process';
 import { spawn, execSync } from 'child_process';
 import { promisify } from 'util';
@@ -83,6 +86,37 @@ import {
 import { createAgentsReadme, createSessionsReadme } from './templates/readme-files.js';
 import { initializeHiveMind, getHiveMindStatus, rollbackHiveMindInit } from './hive-mind-init.js';
 
+// Get the directory path of this module for resolving template files
+const __filename = fileURLToPath(import.meta.url);
+const __dirname = dirname(__filename);
+
+/**
+ * Read the CLAUDE.md template file
+ * @returns {Promise<string>} The template content
+ */
+async function readClaudeMdTemplate() {
+  // In source: templates/CLAUDE.md
+  // In dist: the files are copied directly to init/ directory (not templates/)
+  const possiblePaths = [
+    join(__dirname, 'templates', 'CLAUDE.md'),  // Source location
+    join(__dirname, 'CLAUDE.md'),               // Dist location (files copied directly)
+  ];
+
+  for (const templatePath of possiblePaths) {
+    try {
+      const content = await fs.readFile(templatePath, 'utf8');
+      return content;
+    } catch (error) {
+      // Try next path
+      continue;
+    }
+  }
+
+  // Fallback to generating if template file is not found in any location
+  console.warn('Warning: Template file not found in any location, using generated content');
+  return createOptimizedSparcClaudeMd();
+}
+
 /**
  * Check if Claude Code CLI is installed
  */
@@ -1313,12 +1347,13 @@ async function enhancedClaudeFlowInit(flags, subArgs = []) {
       return;
     }
 
-    // Create CLAUDE.md
+    // Create CLAUDE.md from template
     if (!dryRun) {
-      await fs.writeFile(`${workingDir}/CLAUDE.md`, createOptimizedSparcClaudeMd(), 'utf8');
-      printSuccess('✓ Created CLAUDE.md (Claude Flow v2.0.0 - Optimized)');
+      const claudeMdContent = await readClaudeMdTemplate();
+      await fs.writeFile(`${workingDir}/CLAUDE.md`, claudeMdContent, 'utf8');
+      printSuccess('✓ Created CLAUDE.md (Claude Flow v2.0.0 - Optimized Template)');
     } else {
-      console.log('[DRY RUN] Would create CLAUDE.md (Claude Flow v2.0.0 - Optimized)');
+      console.log('[DRY RUN] Would create CLAUDE.md (Claude Flow v2.0.0 - Optimized Template)');
     }
 
     // Create .claude directory structure

package/src/cli/simple-commands/init/templates/CLAUDE.md

@@ -177,14 +177,12 @@ claude mcp add claude-flow-novice npx claude-flow-novice mcp start
 - `/fullstack "goal"` - Launch full-stack development team with consensus validation
 - `/swarm`, `/sparc`, `/hooks` - Other slash commands (auto-discovered)
 
-## DEVELOPMENT FLOW 
-  1. Execute - Primary swarm (3-8 agents) produces deliverables with confidence score
-  1a. Only when the swarm believes its done move to step 2. 
-  1b. If swarm does not believe it's done (confidence scores < 75%, repeat step 1)
-  2. Verify - Consensus swarm (2-4 validators) runs comprehensive checks with Byzantine voting
-  3. Decision - PASS (≥90% agreement + critical criteria) OR FAIL
-  4. Action -
+## DEVELOPMENT FLOW LOOP
+  1. Execute - Primary swarm (3-8 agents) produces deliverables with confidence score. Each agent MUST be using the enhanced post edit pipeline after file edits.
+  2. When the swarm believes its done with all tasks, move to step 4
+  3. If swarm does not believe it's done (confidence scores < 75%, relaunch agents for step 1)
+  4. Verify - Consensus swarm (2-4 validators) runs comprehensive checks with Byzantine voting
+  5. Decision - PASS (≥90% agreement + critical criteria) OR FAIL
+  6. Action -
     - PASS → Store results → Move to next task
-    - FAIL → Round++ → If <10: inject feedback → Relaunch swarm | If ≥10: Escalate to human
-  5. Repeat - Iterative improvement with accumulated context from all previous rounds
-  6. Escalate - Full history + feedback + recommendations after 10 rounds
+    - FAIL → Round++ → If <10: inject feedback → Relaunch swarm on step 1 | If ≥10: Escalate to human

package/src/slash-commands/benchmark-prompts.js

@@ -0,0 +1,281 @@
+const path = require('path');
+const { AgentBenchmarkSystem } = require('../../benchmark/agent-benchmarking/index');
+
+/**
+ * Benchmark agent prompt formats
+ *
+ * @command /benchmark-prompts
+ * @description Run automated benchmarks comparing different agent prompt formatting styles
+ * @example /benchmark-prompts run
+ * @example /benchmark-prompts run --rounds 5 --parallel
+ * @example /benchmark-prompts analyze
+ * @example /benchmark-prompts report markdown
+ */
+async function benchmarkPrompts(args) {
+  const system = new AgentBenchmarkSystem();
+
+  const command = args[0] || 'help';
+
+  try {
+    switch (command) {
+      case 'run':
+        return await runBenchmark(system, args);
+
+      case 'test':
+        return await testFormat(system, args);
+
+      case 'analyze':
+        return await analyzeBenchmark(system);
+
+      case 'report':
+        return await generateReport(system, args);
+
+      case 'compare':
+        return await compareBenchmark(system);
+
+      case 'list':
+        return await listBenchmarks(system);
+
+      case 'reset':
+        return await resetBenchmarks(system);
+
+      case 'help':
+      default:
+        return showHelp();
+    }
+  } catch (error) {
+    console.error(`\n❌ Error: ${error.message}\n`);
+    if (args.includes('--verbose')) {
+      console.error(error.stack);
+    }
+    return { success: false, error: error.message };
+  }
+}
+
+async function runBenchmark(system, args) {
+  console.log('\n🎯 Running Agent Prompt Format Benchmark\n');
+
+  const options = {
+    rounds: 3,
+    parallel: false,
+    verbose: false,
+    scenarios: null
+  };
+
+  // Parse options
+  for (let i = 1; i < args.length; i++) {
+    const arg = args[i];
+    if (arg === '--rounds' && args[i + 1]) {
+      options.rounds = parseInt(args[i + 1]);
+      i++;
+    } else if (arg === '--parallel') {
+      options.parallel = true;
+    } else if (arg === '--verbose') {
+      options.verbose = true;
+    } else if (arg === '--scenarios' && args[i + 1]) {
+      options.scenarios = args[i + 1];
+      i++;
+    }
+  }
+
+  console.log('Configuration:');
+  console.log(`  Rounds: ${options.rounds}`);
+  console.log(`  Parallel: ${options.parallel ? 'Yes' : 'No'}`);
+  console.log(`  Scenarios: ${options.scenarios || 'All'}\n`);
+
+  const result = await system.run(options);
+
+  console.log('\n✅ Benchmark complete!');
+  console.log('\nNext steps:');
+  console.log('  /benchmark-prompts analyze   - View statistical analysis');
+  console.log('  /benchmark-prompts report    - Generate detailed reports');
+  console.log('  /benchmark-prompts compare   - Interactive comparison\n');
+
+  return { success: true, result };
+}
+
+async function testFormat(system, args) {
+  const format = args[1];
+  if (!format || !['minimal', 'metadata', 'code-heavy'].includes(format)) {
+    console.error('\n❌ Invalid format. Use: minimal, metadata, or code-heavy\n');
+    return { success: false };
+  }
+
+  console.log(`\n🧪 Testing ${format} format\n`);
+
+  // Run benchmark with only specified format
+  const options = {
+    rounds: parseInt(args[2]) || 3,
+    parallel: false,
+    verbose: args.includes('--verbose')
+  };
+
+  // This would need modification to support single format testing
+  console.log('Note: Currently runs all formats. Single format testing coming soon.\n');
+
+  const result = await system.run(options);
+  return { success: true, result };
+}
+
+async function analyzeBenchmark(system) {
+  console.log('\n📊 Analyzing Benchmark Results\n');
+
+  const result = await system.analyze();
+
+  if (!result) {
+    return { success: false };
+  }
+
+  console.log('\n✅ Analysis complete\n');
+  return { success: true, result };
+}
+
+async function generateReport(system, args) {
+  const format = args[1] || 'all';
+
+  if (!['all', 'markdown', 'csv', 'json'].includes(format)) {
+    console.error('\n❌ Invalid format. Use: all, markdown, csv, or json\n');
+    return { success: false };
+  }
+
+  console.log(`\n📄 Generating ${format} report(s)\n`);
+
+  const result = await system.generateReport(format);
+
+  if (!result) {
+    return { success: false };
+  }
+
+  console.log('\n✅ Report generation complete\n');
+  return { success: true, result };
+}
+
+async function compareBenchmark(system) {
+  console.log('\n📊 Interactive Benchmark Comparison\n');
+
+  // Load and analyze latest results
+  const result = await system.analyze();
+
+  if (!result) {
+    return { success: false };
+  }
+
+  const { results, analysis } = result;
+
+  // Print comparison table
+  console.log('Format Comparison:');
+  console.log('─'.repeat(80));
+  console.log('Format'.padEnd(15) + 'Quality'.padEnd(15) + 'Speed'.padEnd(15) + 'Consistency'.padEnd(15) + 'Success');
+  console.log('─'.repeat(80));
+
+  for (const [formatName, formatData] of Object.entries(results.formats)) {
+    const agg = formatData.aggregated;
+    console.log(
+      formatName.padEnd(15) +
+      `${agg.overallQuality.toFixed(1)}%`.padEnd(15) +
+      `${agg.overallResponseTime.toFixed(0)}ms`.padEnd(15) +
+      `${agg.overallConsistency.toFixed(1)}%`.padEnd(15) +
+      `${agg.successRate.toFixed(1)}%`
+    );
+  }
+  console.log('─'.repeat(80));
+
+  console.log('\n🏆 Winner: ' + results.summary.winner.toUpperCase());
+  console.log('\nRecommendations:');
+  for (const rec of results.summary.recommendation) {
+    console.log(`  • ${rec}`);
+  }
+  console.log('');
+
+  return { success: true, results, analysis };
+}
+
+async function listBenchmarks(system) {
+  console.log('\n📂 Benchmark Results\n');
+  await system.listResults();
+  console.log('');
+  return { success: true };
+}
+
+async function resetBenchmarks(system) {
+  console.log('\n⚠️  This will delete all benchmark results.\n');
+
+  // In interactive mode, would ask for confirmation
+  // For now, proceeding with reset
+
+  await system.reset();
+  console.log('');
+  return { success: true };
+}
+
+function showHelp() {
+  console.log(`
+╔════════════════════════════════════════════════════════════════════════════╗
+║                  Agent Prompt Format Benchmark System                      ║
+╚════════════════════════════════════════════════════════════════════════════╝
+
+Compare different agent prompt formatting styles:
+  • Minimal Format    - Simple frontmatter only
+  • Metadata Format   - Full configuration with hooks and capabilities
+  • Code-Heavy Format - Extensive code examples
+
+COMMANDS:
+  /benchmark-prompts run [options]    Run full benchmark suite
+  /benchmark-prompts test <format>    Test specific format
+  /benchmark-prompts analyze          Analyze collected results
+  /benchmark-prompts report [format]  Generate reports (all|markdown|csv|json)
+  /benchmark-prompts compare          Interactive comparison view
+  /benchmark-prompts list             List all benchmark results
+  /benchmark-prompts reset            Clear all benchmark data
+
+OPTIONS:
+  --rounds <n>         Number of test rounds (default: 3)
+  --parallel           Run tests in parallel
+  --scenarios <list>   Comma-separated scenario IDs to test
+  --verbose            Detailed output
+  --export             Export results
+
+EXAMPLES:
+  /benchmark-prompts run
+  /benchmark-prompts run --rounds 5 --parallel
+  /benchmark-prompts test minimal --verbose
+  /benchmark-prompts analyze
+  /benchmark-prompts report markdown
+  /benchmark-prompts compare
+
+TEST SCENARIOS:
+  • simple-code-analysis          - Basic algorithm optimization
+  • memory-leak-detection         - Memory leak identification
+  • database-query-optimization   - Query performance optimization
+  • caching-strategy              - Cache architecture design
+  • resource-allocation           - Resource calculation
+  • async-pattern-optimization    - Async code optimization
+  • algorithm-complexity-reduction - Algorithm improvement
+  • load-testing-strategy         - Load test planning
+  • bottleneck-identification     - Performance bottleneck analysis
+  • scalability-architecture      - Scalability design
+
+BENCHMARK METRICS:
+  • Quality Score    - Completeness, accuracy, relevance, clarity
+  • Response Time    - Total time to generate response
+  • Consistency      - Variance across multiple rounds
+  • Success Rate     - Percentage of successful completions
+
+For more information, see: /benchmark/agent-benchmarking/README.md
+`);
+
+  return { success: true };
+}
+
+module.exports = {
+  command: 'benchmark-prompts',
+  description: 'Run automated benchmarks comparing agent prompt formats',
+  handler: benchmarkPrompts,
+  examples: [
+    '/benchmark-prompts run',
+    '/benchmark-prompts run --rounds 5 --parallel',
+    '/benchmark-prompts analyze',
+    '/benchmark-prompts report markdown',
+    '/benchmark-prompts compare'
+  ]
+};