honest-drive-mcp

honest-drive-mcp

A local Google Drive MCP server with full permission management (share, list, remove permissions), enabling AI agents to manage file sharing and access beyond the official connector.

Category
访问服务器

README

honest-drive-mcp

Local Google Drive MCP server with full permission management — the piece missing from the official connector.

Companion project to honest-gmail-mcp and honest-calendar-mcp.

Why

The official Anthropic Drive connector is competent for read + create, but has one big gap: it cannot set or modify sharing permissions. If you build a workflow like "create a folder for my accountant and share it with her", you hit a dead end — you have to jump into the Drive UI to do the sharing part manually.

This MCP fixes that. Same trust model as the rest of the honest-* family: data flows only between your machine and Google.

Features

Eight tools exposed over MCP:

  • list_files — search with Drive query syntax
  • get_file — full metadata
  • read_file — download content (text/JSON returned as UTF-8; binaries as base64; Google-native docs auto-exported to plain text / CSV)
  • create_file — upload from text or base64
  • create_folder — new folder, optionally inside a parent
  • share_file — grant access to an email (reader/commenter/writer/owner), notification optional, custom message supported ← the piece missing elsewhere
  • list_permissions — see who has access and with what role
  • remove_permission — revoke a specific permission by id

Requirements

  • Python 3.10+
  • A Google account
  • One-time Google Cloud OAuth setup (~10 min; can reuse the client from honest-gmail-mcp — just enable the Drive API on the same project)

Setup

git clone https://github.com/bartosz-kuc/honest-drive-mcp.git
cd honest-drive-mcp
python3 -m venv venv
./venv/bin/pip install -r requirements.txt

Google Cloud steps (same as honest-gmail-mcp):

  1. https://console.cloud.google.com/ → create or select project
  2. APIs & Services → Library → enable Google Drive API
  3. APIs & Services → Credentials → OAuth client ID → Desktop app → download JSON → save as credentials.json here

First run:

./venv/bin/python server.py

Browser opens → Allow → token saved to token.json. Ctrl+C.

Register with Claude Code:

claude mcp add drive-personal /absolute/path/to/venv/bin/python /absolute/path/to/server.py

Claude Desktop claude_desktop_config.json:

{
  "mcpServers": {
    "drive-personal": {
      "command": "/absolute/path/to/venv/bin/python",
      "args": ["/absolute/path/to/server.py"]
    }
  }
}

Example usage

"Create a folder called 'Faktury dla księgowej' in My Drive and share it with elwira@example.pl as writer."

AI calls create_folder → gets folder_id → calls share_file(file_id, "elwira@example.pl", role="writer"). Done in one shot. No context-switch to the Drive UI.

"Who has access to file X?"

AI calls list_permissions(file_id) → returns all permissions with ids you can pass to remove_permission later.

Data flow

Your AI client
     ↕  MCP stdio
This server (Python, on your machine)
     ↕  HTTPS
Google Drive API

No cloud middle. Tokens stay on disk, .gitignored.

Security notes

  • OAuth scope requested: drive (full read/write on all Drive files this account can access). Google does not offer a scope that includes permission management without also including read/write to files, which is why the scope is broad.
  • send_notification defaults to false — AI will not spam recipients unless explicitly asked.
  • role: "owner" transfers ownership — an irreversible action. The tool passes transferOwnership: true automatically when this role is requested; consider using writer unless you really mean to hand over the file.
  • Revoke access anytime at https://myaccount.google.com/permissions.
  • credentials.json and token.json are .gitignored.

Author

Bartosz Kuć — Warsaw-based developer, JDG owner running skanfirmy.pl.

  • Site: https://skanfirmy.pl

  • GitHub: https://github.com/bartosz-kuc

  • Email: firma@bartosza.pl

Consulting

Available for consulting on Polish tax and business integrations (KSeF, GUS/NFZ/GIOŚ APIs, mBank data), MCP server design, and AI-assisted tooling for JDGs and small teams. See skanfirmy.pl/uslugi for productized packages (audit 3k PLN, setup 8-15k PLN, retainer 2-4k PLN/mo), or reach out via email.

License

MIT — see LICENSE.

Related

推荐服务器

Baidu Map

Baidu Map

百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。

官方
精选
JavaScript
Playwright MCP Server

Playwright MCP Server

一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。

官方
精选
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。

官方
精选
本地
TypeScript
VeyraX

VeyraX

一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。

官方
精选
本地
graphlit-mcp-server

graphlit-mcp-server

模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。

官方
精选
TypeScript
Kagi MCP Server

Kagi MCP Server

一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

Exa MCP Server

模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。

官方
精选