metona-ai-desktop/electron/harness/tools/built-in/filesystem.ts

/**
 * 文件系统工具（4 个）
 *
 * read_file, write_file, list_directory, search_files
 *
 * 安全策略：
 * 1. 路径遍历防护 — 所有路径必须解析到工作空间内（修复前缀碰撞漏洞）
 * 2. 受保护文件拦截 — MEMORY.md 仅由系统内部 WorkspaceService 管理，
 *    禁止任何文件工具直接读写（仅限工作空间根目录，子目录不受限）
 *
 * @see docs/MetonaAI-Desktop 架构与交互设计.html — 9 个基础工具
 * @see standard/开发规范.md — 使用 fs/path 内置模块
 */

import { readFile, writeFile, readdir, stat, appendFile, mkdir, open } from 'fs/promises';
import { join, relative, resolve, dirname } from 'path';
import { existsSync } from 'fs';
import type { IMetonaTool, ToolExecutionContext } from '../../types/metona-tool';
import type { MetonaToolDef } from '../../../harness/types';
import { MetonaToolCategory, MetonaRiskLevel } from '../../../harness/types';
import { isProtectedWorkspaceFile, isPathWithinWorkspace } from './file-guard';

/** 共享路径解析 + 安全校验 */
function safeResolvePath(filePath: string, workspacePath: string): string {
  const resolved = resolve(workspacePath, filePath);
  // 安全检查：路径遍历防护（修复前缀碰撞漏洞）
  if (!isPathWithinWorkspace(filePath, workspacePath)) {
    throw new Error(`Path traversal detected: ${filePath}`);
  }
  // 受保护文件检查：MEMORY.md 仅由系统内部管理
  if (isProtectedWorkspaceFile(filePath, workspacePath)) {
    throw new Error('Access denied: MEMORY.md is managed by the memory system and cannot be accessed via file tools');
  }
  return resolved;
}

/** 共享 glob 匹配（简易通配符 → 正则） */
function matchGlob(name: string, glob: string): boolean {
  const pattern = glob.replace(/[.+^${}()|[\]\\]/g, '\\$&').replace(/\*/g, '.*').replace(/\?/g, '.');
  return new RegExp(`^${pattern}$`, 'i').test(name);
}

/** 二进制文件检测：读取前 8KB 检查是否含 NULL 字节 */
async function isBinaryFile(filePath: string): Promise<boolean> {
  try {
    const buffer = Buffer.alloc(8192);
    const fd = await open(filePath, 'r');
    await fd.read(buffer, 0, 8192, 0);
    await fd.close();
    // 含 NULL 字节 → 二进制
    for (let i = 0; i < buffer.length; i++) {
      if (buffer[i] === 0) return true;
    }
    return false;
  } catch {
    return false;
  }
}

// ===== 1. read_file =====

export class ReadFileTool implements IMetonaTool {
  readonly definition: MetonaToolDef = {
    name: 'read_file',
    description: 'Read the contents of a file. Returns the full text content. Supports line offset and limit for large files.',
    parameters: {
      type: 'object',
      properties: {
        file_path: { type: 'string', description: 'Absolute or relative path to the file to read' },
        offset: { type: 'number', description: 'Start line number (1-indexed, default 1)' },
        limit: { type: 'number', description: 'Maximum lines to read (default 500, max 2000)' },
      },
      required: ['file_path'],
    },
    category: MetonaToolCategory.FILESYSTEM,
    riskLevel: MetonaRiskLevel.SAFE,
    requiresPermission: false,
    timeoutMs: 10_000,
  };

  async execute(args: Record<string, unknown>, context: ToolExecutionContext): Promise<unknown> {
    const filePath = safeResolvePath(args.file_path as string, context.workspacePath);
    const offset = Math.max(1, (args.offset as number) ?? 1);
    const limit = Math.min(2000, Math.max(1, (args.limit as number) ?? 500));

    // 二进制文件检测 — 避免读取图片/可执行文件产生乱码
    if (await isBinaryFile(filePath)) {
      return {
        content: '',
        total_lines: 0,
        returned_lines: 0,
        truncated: false,
        file_size: (await stat(filePath)).size,
        error: 'Binary file detected. Use web_browser screenshot or other tools for binary content.',
      };
    }

    const content = await readFile(filePath, 'utf-8');
    const lines = content.split('\n');
    const slicedLines = lines.slice(offset - 1, offset - 1 + limit);

    return {
      content: slicedLines.join('\n'),
      total_lines: lines.length,
      returned_lines: slicedLines.length,
      truncated: lines.length > offset - 1 + limit,
      file_size: Buffer.byteLength(content, 'utf-8'),
    };
  }
}

// ===== 2. write_file =====

export class WriteFileTool implements IMetonaTool {
  readonly definition: MetonaToolDef = {
    name: 'write_file',
    description: 'Write content to a file. Creates the file if it doesn\'t exist, overwrites if it does. Supports append mode.',
    parameters: {
      type: 'object',
      properties: {
        file_path: { type: 'string', description: 'Path to the file to write' },
        content: { type: 'string', description: 'Content to write to the file' },
        mode: { type: 'string', description: 'Write mode: "overwrite" (default) or "append"', enum: ['overwrite', 'append'] },
      },
      required: ['file_path', 'content'],
    },
    category: MetonaToolCategory.FILESYSTEM,
    riskLevel: MetonaRiskLevel.MEDIUM,
    requiresPermission: true,
    timeoutMs: 15_000,
  };

  async execute(args: Record<string, unknown>, context: ToolExecutionContext): Promise<unknown> {
    const filePath = safeResolvePath(args.file_path as string, context.workspacePath);
    const content = args.content as string;
    const mode = (args.mode as string) ?? 'overwrite';

    // 自动创建父目录（递归）
    const parentDir = dirname(filePath);
    if (!existsSync(parentDir)) {
      await mkdir(parentDir, { recursive: true });
    }

    if (mode === 'append') {
      await appendFile(filePath, content, 'utf-8');
    } else {
      await writeFile(filePath, content, 'utf-8');
    }

    return { bytes_written: Buffer.byteLength(content, 'utf-8'), path: filePath };
  }
}

// ===== 3. list_directory =====

export class ListDirectoryTool implements IMetonaTool {
  readonly definition: MetonaToolDef = {
    name: 'list_directory',
    description: 'List the contents of a directory. Supports recursive depth control and glob filtering.',
    parameters: {
      type: 'object',
      properties: {
        dir_path: { type: 'string', description: 'Directory path (default: workspace root)' },
        depth: { type: 'number', description: 'Recursive depth (default 1, max 5)' },
        glob: { type: 'string', description: 'Filename filter pattern (e.g., "*.ts")' },
      },
    },
    category: MetonaToolCategory.FILESYSTEM,
    riskLevel: MetonaRiskLevel.SAFE,
    requiresPermission: false,
    timeoutMs: 10_000,
  };

  async execute(args: Record<string, unknown>, context: ToolExecutionContext): Promise<unknown> {
    const dirPath = args.dir_path
      ? safeResolvePath(args.dir_path as string, context.workspacePath)
      : context.workspacePath;
    const depth = Math.min(5, Math.max(1, (args.depth as number) ?? 1));
    const glob = args.glob as string | undefined;

    const entries = await this.listDir(dirPath, dirPath, depth, glob, 0);
    return { entries, count: entries.length };
  }

  private async listDir(
    rootPath: string,
    dirPath: string,
    maxDepth: number,
    glob: string | undefined,
    currentDepth: number,
  ): Promise<Array<{ name: string; path: string; type: string; size?: number }>> {
    const results: Array<{ name: string; path: string; type: string; size?: number }> = [];

    try {
      const entries = await readdir(dirPath, { withFileTypes: true });

      for (const entry of entries) {
        // 跳过隐藏文件和 node_modules
        if (entry.name.startsWith('.') || entry.name === 'node_modules') continue;

        const fullPath = join(dirPath, entry.name);
        // 相对路径始终以根请求目录为基准
        const relativePath = relative(rootPath, fullPath);

        if (entry.isDirectory()) {
          // 目录始终列出（不受 glob 过滤），保证递归可进入子目录
          results.push({ name: entry.name, path: relativePath, type: 'directory' });
          if (currentDepth < maxDepth - 1) {
            const subEntries = await this.listDir(rootPath, fullPath, maxDepth, glob, currentDepth + 1);
            results.push(...subEntries);
          }
        } else {
          // glob 过滤仅适用于文件
          if (glob && !matchGlob(entry.name, glob)) continue;
          const stats = await stat(fullPath);
          results.push({ name: entry.name, path: relativePath, type: 'file', size: stats.size });
        }
      }
    } catch {
      // 目录读取失败，跳过
    }

    return results;
  }
}

// ===== 4. search_files =====

export class SearchFilesTool implements IMetonaTool {
  readonly definition: MetonaToolDef = {
    name: 'search_files',
    description: 'Search for files by name pattern or content regex. Uses efficient file scanning.',
    parameters: {
      type: 'object',
      properties: {
        pattern: { type: 'string', description: 'Search pattern (regex for content, glob for filenames)' },
        target: { type: 'string', description: '"content" (default) to search file contents, "files" to search filenames', enum: ['content', 'files'] },
        path: { type: 'string', description: 'Search directory (default: workspace root)' },
        file_glob: { type: 'string', description: 'Limit to specific file types (e.g., "*.py")' },
        limit: { type: 'number', description: 'Maximum results (default 50)' },
      },
      required: ['pattern'],
    },
    category: MetonaToolCategory.FILESYSTEM,
    riskLevel: MetonaRiskLevel.SAFE,
    requiresPermission: false,
    timeoutMs: 30_000,
  };

  async execute(args: Record<string, unknown>, context: ToolExecutionContext): Promise<unknown> {
    const pattern = args.pattern as string;
    const target = (args.target as string) ?? 'content';
    const searchPath = args.path
      ? this.resolveSearchPath(args.path as string, context.workspacePath)
      : context.workspacePath;
    const fileGlob = args.file_glob as string | undefined;
    const limit = Math.min(200, Math.max(1, (args.limit as number) ?? 50));

    if (target === 'files') {
      return this.searchByFilename(searchPath, pattern, fileGlob, limit);
    }
    return this.searchByContent(searchPath, pattern, fileGlob, limit, context.workspacePath);
  }

  private async searchByFilename(dirPath: string, pattern: string, fileGlob: string | undefined, limit: number): Promise<unknown> {
    const results: Array<{ path: string; name: string }> = [];

    await this.walkDir(dirPath, async (filePath, name) => {
      if (results.length >= limit) return;
      if (matchGlob(name, pattern)) {
        results.push({ path: relative(dirPath, filePath), name });
      }
    }, fileGlob);

    return { results, count: results.length };
  }

  private async searchByContent(dirPath: string, pattern: string, fileGlob: string | undefined, limit: number, workspacePath: string): Promise<unknown> {
    const results: Array<{ path: string; line: number; match: string }> = [];

    // 正则安全加固：限制 pattern 长度 + try-catch 防止 ReDoS
    if (pattern.length > 500) {
      return { results: [], count: 0, error: 'Search pattern too long (max 500 chars)' };
    }
    let regex: RegExp;
    try {
      regex = new RegExp(pattern, 'gi');
    } catch {
      return { results: [], count: 0, error: `Invalid regex pattern: ${pattern}` };
    }

    await this.walkDir(dirPath, async (filePath) => {
      if (results.length >= limit) return;
      // 跳过工作空间根目录的 MEMORY.md（受保护文件，使用完整路径精确匹配）
      if (isProtectedWorkspaceFile(filePath, workspacePath)) {
        return;
      }
      try {
        const content = await readFile(filePath, 'utf-8');
        const lines = content.split('\n');
        for (let i = 0; i < lines.length; i++) {
          if (results.length >= limit) break;
          if (regex.test(lines[i])) {
            results.push({
              path: relative(dirPath, filePath),
              line: i + 1,
              match: lines[i].trim().slice(0, 200),
            });
          }
          regex.lastIndex = 0;
        }
      } catch {
        // 跳过不可读文件
      }
    }, fileGlob);

    return { results, count: results.length };
  }

  private async walkDir(
    dirPath: string,
    callback: (filePath: string, name: string) => Promise<void>,
    fileGlob?: string,
  ): Promise<void> {
    try {
      const entries = await readdir(dirPath, { withFileTypes: true });
      for (const entry of entries) {
        if (entry.name.startsWith('.') || entry.name === 'node_modules') continue;

        const fullPath = join(dirPath, entry.name);
        if (entry.isDirectory()) {
          await this.walkDir(fullPath, callback, fileGlob);
        } else {
          if (fileGlob && !matchGlob(entry.name, fileGlob)) continue;
          await callback(fullPath, entry.name);
        }
      }
    } catch {
      // 目录读取失败
    }
  }

  /** search_files 的路径解析：仅需遍历防护，不需要 MEMORY.md 拦截（搜索时单独跳过） */
  private resolveSearchPath(filePath: string, workspacePath: string): string {
    const resolved = resolve(workspacePath, filePath);
    if (!isPathWithinWorkspace(filePath, workspacePath)) {
      throw new Error(`Path traversal detected: ${filePath}`);
    }
    return resolved;
  }
}