my-vault/.scripts/verify-vault.mjs

#!/usr/bin/env node

import { execSync } from 'node:child_process'
import { readFileSync } from 'node:fs'

// 值形态的占位符 / 环境引用：不算泄漏
const PLACEHOLDER =
  /(\{\{|<[^>]*>|\$\{|process\.env|\benv\[|your[-_]|example|sample|placeholder|changeme|redacted|\bxxx\b|\.\.\.|\[hidden\])/i

// 只报路径与行号，绝不打印匹配到的内容
const KEY_PATTERNS = [
  // OpenAI 系（sk-or-v1 / sk-proj / sk-svcacct / 裸 sk-），键内允许连字符
  ['openai-key', /\bsk-[A-Za-z0-9_-]{16,}/, false],
  // 第三方 "xxxsk-" 前缀族（如 ctx7sk-…）
  ['prefixed-key', /\b[A-Za-z0-9]{2,8}sk-[A-Za-z0-9_-]{16,}/, false],
  ['aws', /\bAKIA[0-9A-Z]{16}\b/, false],
  ['github-pat', /\bgh[pousr]_[A-Za-z0-9]{20,}/, false],
  ['gitlab-pat', /\bglpat-[A-Za-z0-9_-]{20,}/, false],
  // 覆盖 OPENSSH / RSA / EC / ENCRYPTED 等全部变体
  ['private-key', /BEGIN [A-Z0-9 ]*PRIVATE KEY/, false],
  ['slack', /\bxox[baprs]-[A-Za-z0-9-]{10,}/, false],
  // 关键字赋值 / 表格单元格：PASSWORD=… / token: … / | API_KEY | …（值须 ≥16 位密钥形态字符）
  [
    'credential-assign',
    /[A-Za-z0-9_-]{0,32}(?:password|passwd|pwd|token|secret|api[_-]?key|apikey|access[_-]?key|secret[_-]?key|bot[_-]?token|private[_-]?key|credential)["']?\s*[:=|]\s*["']?([A-Za-z0-9_+/=~.!@#$%^&*?:-]{16,})/i,
    true,
  ],
  // Telegram bot token：<数字id>:AA<hash>
  ['telegram-bot', /\b\d{8,10}:AA[A-Za-z0-9_-]{30,}/, false],
  // 带内嵌口令的连接串 postgres://user:pass@host
  [
    'conn-string',
    /\b(?:postgres(?:ql)?|mysql|redis|mongodb(?:\+srv)?|amqp|smtp|mssql):\/\/[^\s'"@:/]+:[^\s'"@]{4,}@/,
    false,
  ],
]

const OBSIDIAN_HEX_PATTERN = /^[0-9a-f]{64}$/i

let found = false

const trackedFiles = execSync('git ls-files', { encoding: 'utf8' })
  .split('\n')
  .filter(Boolean)

for (const file of trackedFiles) {
  if (file === '.scripts/verify-vault.mjs') continue

  if (file.endsWith('.png') || file.endsWith('.jpg') || file.endsWith('.jpeg') || file.endsWith('.gif') || file.endsWith('.webp') || file.endsWith('.svg') || file.endsWith('.ico') || file.endsWith('.woff') || file.endsWith('.woff2') || file.endsWith('.ttf') || file.endsWith('.eot') || file.endsWith('.pdf') || file.endsWith('.zip') || file.endsWith('.gz') || file.endsWith('.tar') || file.endsWith('.mp4') || file.endsWith('.mp3') || file.endsWith('.wav')) {
    continue
  }

  let content
  try {
    content = readFileSync(file, 'utf8')
  } catch {
    continue
  }

  const lines = content.split('\n')
  // .obsidian/** 是第三方插件代码与状态，值形态规则误报极高（minified JS/CSS）；
  // 密钥载体 data.json 已解除跟踪，这里只保留 64 位 hex 检查
  const isObsidian = file.startsWith('.obsidian/')
  for (let i = 0; i < lines.length; i++) {
    const line = lines[i]

    if (!isObsidian) {
      for (const [name, pattern, checkValue] of KEY_PATTERNS) {
        const m = pattern.exec(line)
        if (!m) continue
        // 赋值类：值是占位符 / 环境引用时放行
        if (checkValue) {
          const value = m[1]
          if (!value || PLACEHOLDER.test(value)) continue
        }
        console.error(`KEY LEAK (${name}): ${file}:${i + 1}`)
        found = true
      }
    }

    if (file.startsWith('.obsidian/plugins/') && file.endsWith('.json')) {
      const tokens = line.split(/[\s":,]+/).filter(Boolean)
      for (const token of tokens) {
        if (OBSIDIAN_HEX_PATTERN.test(token) && token.length === 64) {
          console.error(`OBSIDIAN HEX KEY: ${file}:${i + 1}`)
          found = true
          break
        }
      }
    }
  }
}

if (found) {
  process.exit(1)
}