obsidian-mcp/src/validation/sanitizer.ts

/**
 * Parameter Sanitizer
 * Constitutional Principle III: Local Execution Security
 * Remove dangerous characters and validate inputs before CLI execution
 */

import { logger } from '../utils/logger.js';

/**
 * Characters that should be removed or escaped for security
 * Note: Brackets [], parentheses (), and braces {} are safe because values are quoted and passed as array args
 * They're essential for Obsidian markdown (wikilinks [[link]], tasks - [ ] Task, templates {{...}}, etc.)
 * Note: Single & is safe in quoted args (filenames like "Research & Development.md")
 * Note: Backticks are safe because formatParam escapes them as \` inside double-quoted strings,
 *       preventing shell command substitution while preserving Markdown code fences (``` ```)
 * Note: < and > are safe inside double-quoted strings — shell redirection only applies at the
 *       command level, not inside quotes. Stripping them breaks Mermaid arrows (->>, -->) and HTML.
 * We only block: ; | $ (command separators, pipes, variable substitution)
 * Command injection patterns (&&, ||, etc.) are handled separately
 */
const DANGEROUS_CHARS = /[;|$]/g;
const COMMAND_INJECTION_PATTERNS = [
  /\$\(/g,      // Command substitution $(...)
  /\|\|/g,      // OR operator
  /&&/g,        // AND operator
  /;/g,         // Command separator
];

/**
 * Sanitize a single string parameter
 */
export function sanitizeString(input: string): string {
  if (typeof input !== 'string') {
    logger.warn('sanitizeString received non-string input', { type: typeof input });
    return String(input);
  }

  // Remove null bytes
  let sanitized = input.replace(/\0/g, '');

  // Check for command injection patterns
  for (const pattern of COMMAND_INJECTION_PATTERNS) {
    if (pattern.test(sanitized)) {
      logger.warn('Potential command injection detected', {
        pattern: pattern.toString(),
      });
      sanitized = sanitized.replace(pattern, '');
    }
  }

  // Remove dangerous characters
  sanitized = sanitized.replace(DANGEROUS_CHARS, '');

  // Trim whitespace
  sanitized = sanitized.trim();

  return sanitized;
}

/**
 * Sanitize a file path
 */
export function sanitizePath(path: string): string {
  if (typeof path !== 'string') {
    return '';
  }

  // Remove null bytes
  let sanitized = path.replace(/\0/g, '');

  // Remove path traversal attempts
  sanitized = sanitized.replace(/\.\./g, '');

  // Remove leading/trailing slashes (relative paths only)
  sanitized = sanitized.replace(/^\/+|\/+$/g, '');

  // Remove dangerous characters but allow path separators
  // Note: Brackets, parentheses, braces, and single & are safe in paths (quoted args)
  // Note: < and > are safe inside double-quoted strings (not shell redirects)
  sanitized = sanitized.replace(/[;|`$]/g, '');

  return sanitized;
}

/**
 * Sanitize a tag (must start with #)
 */
export function sanitizeTag(tag: string): string {
  if (typeof tag !== 'string') {
    return '#';
  }

  let sanitized = tag.trim();

  // Ensure tag starts with #
  if (!sanitized.startsWith('#')) {
    sanitized = '#' + sanitized;
  }

  // Allow only alphanumeric, hyphens, underscores, and forward slashes
  sanitized = sanitized.replace(/[^#a-zA-Z0-9_/-]/g, '');

  // Remove empty tags
  if (sanitized === '#') {
    return '';
  }

  return sanitized;
}

/**
 * Sanitize object parameters recursively
 */
export function sanitizeParameters(params: Record<string, unknown>): Record<string, unknown> {
  const sanitized: Record<string, unknown> = {};

  for (const [key, value] of Object.entries(params)) {
    if (typeof value === 'string') {
      // Special handling for different parameter types
      if (key === 'tag' || key.startsWith('tag')) {
        sanitized[key] = sanitizeTag(value);
      } else if (key === 'path' || key.endsWith('Path') || key.endsWith('path')) {
        sanitized[key] = sanitizePath(value);
      } else {
        sanitized[key] = sanitizeString(value);
      }
    } else if (typeof value === 'object' && value !== null && !Array.isArray(value)) {
      // Recursively sanitize nested objects
      sanitized[key] = sanitizeParameters(value as Record<string, unknown>);
    } else if (Array.isArray(value)) {
      // Sanitize array elements
      sanitized[key] = value.map((item) =>
        typeof item === 'string' ? sanitizeString(item) : item
      );
    } else {
      // Keep other types as-is (numbers, booleans, null)
      sanitized[key] = value;
    }
  }

  return sanitized;
}

/**
 * Validate that parameters don't contain shell metacharacters
 */
export function containsDangerousCharacters(input: string): boolean {
  if (typeof input !== 'string') {
    return false;
  }

  // Check for dangerous characters
  if (DANGEROUS_CHARS.test(input)) {
    return true;
  }

  // Check for command injection patterns
  for (const pattern of COMMAND_INJECTION_PATTERNS) {
    if (pattern.test(input)) {
      return true;
    }
  }

  return false;
}