tractatus/scripts/check-prohibited-terms.js

#!/usr/bin/env node
/**
 * Prohibited Terms Scanner - Enforces inst_016, inst_017, inst_018
 * 
 * Scans files for prohibited language that violates governance rules
 */

const fs = require('fs');
const path = require('path');
const { execSync } = require('child_process');

// inst_017: Prohibited absolute assurance terms
const PROHIBITED_ABSOLUTE_TERMS = [
  /\bguarantee(s|d)?\b/i,
  /\bensures?\s+(100%|complete|total|absolute)/i,
  /\beliminates?\s+all\b/i,
  /\bcompletely\s+(prevents?|eliminates?|removes?)\b/i,
  /\bnever\s+fails?\b/i,
  /\b100%\s+(safe|secure|reliable|accurate)\b/i,
  /\babsolutely\s+(prevents?|guarantees?)\b/i
];

// inst_018: Prohibited maturity claims without evidence
const PROHIBITED_MATURITY_CLAIMS = [
  /\b(production-ready|battle-tested|enterprise-proven)\b/i,
  /\bvalidated\s+by\s+\d+\s+(companies|organizations|teams)\b/i,
  /\bwidely\s+adopted\b/i,
  /\bmarket\s+(leader|validated)\b/i,
  /\bcustomer\s+base\s+of\b/i
];

// inst_016: Requires citation or [NEEDS VERIFICATION]
const STATS_PATTERNS = [
  /\d+%\s+(improvement|increase|reduction|faster|better)/i,
  /\d+x\s+(faster|better|more)/i,
  /ROI\s+of\s+\d+/i,
  /reduces?\s+(cost|time|effort)\s+by\s+\d+/i
];

function checkFile(filePath) {
  const content = fs.readFileSync(filePath, 'utf8');
  const lines = content.split('\n');
  const violations = [];

  lines.forEach((line, idx) => {
    const lineNum = idx + 1;

    // Check inst_017: Absolute assurance terms
    PROHIBITED_ABSOLUTE_TERMS.forEach(pattern => {
      if (pattern.test(line)) {
        violations.push({
          file: filePath,
          line: lineNum,
          type: 'inst_017',
          severity: 'HIGH',
          text: line.trim(),
          rule: 'Prohibited absolute assurance term detected'
        });
      }
    });

    // Check inst_018: Maturity claims
    PROHIBITED_MATURITY_CLAIMS.forEach(pattern => {
      if (pattern.test(line)) {
        violations.push({
          file: filePath,
          line: lineNum,
          type: 'inst_018',
          severity: 'HIGH',
          text: line.trim(),
          rule: 'Prohibited maturity claim without evidence'
        });
      }
    });

    // Check inst_016: Statistics without citation
    STATS_PATTERNS.forEach(pattern => {
      if (pattern.test(line)) {
        // Check if line has citation or [NEEDS VERIFICATION]
        if (!line.includes('[') && !line.includes('(source:') && !line.includes('Citation:')) {
          violations.push({
            file: filePath,
            line: lineNum,
            type: 'inst_016',
            severity: 'MEDIUM',
            text: line.trim(),
            rule: 'Statistic without citation or [NEEDS VERIFICATION] marker'
          });
        }
      }
    });
  });

  return violations;
}

function scanFiles(files) {
  const allViolations = [];

  files.forEach(file => {
    if (!fs.existsSync(file)) return;
    
    // Skip instruction-history.json (contains prohibited term DEFINITIONS, not violations)
    if (file.includes('instruction-history.json')) return;
    
    // Only scan text files (markdown, HTML, text)
    const ext = path.extname(file).toLowerCase();
    if (!['.md', '.html', '.txt', '.json'].includes(ext)) return;

    const violations = checkFile(file);
    allViolations.push(...violations);
  });

  return allViolations;
}

function main() {
  const args = process.argv.slice(2);

  let files = [];
  if (args.length === 0) {
    // Scan staged git files
    try {
      const staged = execSync('git diff --cached --name-only --diff-filter=ACM', {
        encoding: 'utf8'
      });
      files = staged.trim().split('\n').filter(f => f.length > 0);
    } catch (err) {
      console.error('Not in a git repository or no staged files');
      process.exit(0);
    }
  } else {
    files = args;
  }

  if (files.length === 0) {
    console.log('✅ No files to scan');
    process.exit(0);
  }

  console.log(`\n🔍 Scanning ${files.length} file(s) for prohibited terms...\n`);

  const violations = scanFiles(files);

  if (violations.length === 0) {
    console.log('✅ No prohibited terms detected\n');
    process.exit(0);
  }

  // Report violations
  console.log(`❌ Found ${violations.length} violation(s):\n`);
  console.log('━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n');

  violations.forEach(v => {
    console.log(`${v.severity === 'HIGH' ? '🔴' : '🟡'} ${v.file}:${v.line}`);
    console.log(`   Rule: ${v.type} - ${v.rule}`);
    console.log(`   Text: ${v.text.substring(0, 80)}${v.text.length > 80 ? '...' : ''}`);
    console.log('');
  });

  console.log('━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\n');
  console.log('Fix violations before committing/deploying:\n');
  console.log('  inst_016: Add citation or [NEEDS VERIFICATION] to statistics');
  console.log('  inst_017: Replace absolute terms with evidence-based language');
  console.log('  inst_018: Remove maturity claims or add documented evidence\n');

  process.exit(1);
}

main();