MCP File Organizer
MCP File Organizer — 技术博客文章“MCP in Practice”的演示
这是一个通过“AI File Organization Assistant”用例来演示 MCP 的示例:一个 MCP Server 公开 2 个文件管理工具(list_files、move_file — 没有 read_file;服务器从不读取文件内容,只读取文件名/扩展名/修改日期),以及一个 MCP Client/Agent,它使用 Gemini(通过 @google/genai,借助 mcpToTool 提供内置的 MCP 支持)根据用户的自然语言请求决定调用哪个工具。folder 是机器上的任意绝对路径(不限于演示的 storage/inbox — 它可以指向真实文件夹,例如 Downloads),结果总是落在 <folder>/organized/<extension>/<year>/<month>/<day>/ 中。演示数据集是 500 个在 365 天内累积的模拟文件,先按扩展名、再按最后修改的年/月/日重新组织。
设置
npm installRelated MCP server: Organizer MCP Server
生成模拟数据
npm run generate-mock # generates 500 mock files into storage/inbox, with mtimes spread across the last 365 days随时重新运行此命令即可生成全新的数据集。
运行真实 Agent(需要 Gemini API 密钥)
client/client.ts 是真正的 Agent:它通过 mcpToTool 将 MCP 客户端暴露给 Gemini,让 Gemini 自行决定调用哪个工具以及使用什么参数(自动函数调用),没有任何硬编码规则。它从 GEMINI_API_KEY、GOOGLE_API_KEY 或 API_KEY 中读取密钥(按此优先级顺序);如果 MODEL 的值以 gemini- 开头,则从 MODEL 读取模型,否则回退到默认的 gemini-2.5-flash。
# PowerShell
$env:GEMINI_API_KEY = "AIza..."
npm run agent# bash
export GEMINI_API_KEY="AIza..."
npm run agent这些变量也可以设置在项目目录下的 .env 文件中;npm run agent 会在 .env 存在时自动加载它(使用 Node 的 --env-file-if-exists 标志)。
如果要在真实文件夹(而不是 storage/inbox)上尝试,请传入绝对路径作为参数,例如 npm run agent -- "C:\Users\you\Downloads"。
使用真实数据时的安全性
server.ts 在接触文件系统之前已经针对多种情况做了防护:
folder必须是绝对路径 — 相对路径会立即被拒绝,绝不会被静默地相对于服务器进程的 cwd 解析。name必须是纯文件名(不能包含/、\或..)— 阻止越出folder的路径遍历。to_folder会被解析,然后重新检查以确保它保持在<folder>/organized/内 — 阻止通过..越出。如果目标已存在,
move_file会拒绝执行(绝不会静默覆盖);传入dry_run: true可以预览将要移动的内容,而不会触碰真实文件。设置
ALLOWED_ROOTS环境变量(绝对路径列表,在 Windows 上用;分隔,在 Unix 上用:分隔 — 遵循path.delimiter),将folder限制为仅这些根目录。如果未设置,演示保持原有行为:接受任何绝对文件夹 — 用于演示没问题,但在指向真实数据时应设置ALLOWED_ROOTS。
结构
server/server.ts MCP Server: list_files, move_file — folder is any absolute path, never reads file contents
client/generate-mock-inbox.ts Generates 500 mock files into storage/inbox
client/client.ts MCP Client + Agent, using Gemini via @google/genai
storage/ Generated by running generate-mock, not committed to the repoThis server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Related MCP Servers
- AlicenseNot gradedqualityCmaintenanceEnables LLMs to organize and manage Windows file systems with intelligent file analysis, automated grouping, renaming with date prefixes, and safe operations through dry-run mode and sandbox restrictions.MIT
- AlicenseNot gradedqualityDmaintenanceEnables Claude to intelligently organize files by scanning folders, detecting duplicates, suggesting meaningful names, and moving files into logical folders.MIT
- AlicenseAqualityCmaintenanceEnables advanced file system operations such as file read/write, directory management, search, archiving, and file watching through natural language.16112MIT
- AlicenseNot gradedqualityDmaintenanceEnables file operations such as counting, listing, compressing images, creating/extracting archives, copying/moving files, and merging/splitting PDFs via natural language.5MIT
Related MCP Connectors
Securely search and manage workspace context files for AI agents and teams.
File uploads for AI agents. Upload, list, and manage files. No signup required.
Upload, organize, search, and transform images, videos, and files with AI-powered tools.
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/BWV-Personal-Act/thaipq-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server