Google Calendar Meeting Setup
Enables creating Google Calendar meeting invites with automated authentication via OAuth. Supports scheduling meetings with customizable title, duration, notes, and attendees through a simple command-line or MCP tool interface.
README
Google Calendar Meeting Setup
Automation script for creating Google Calendar meeting invites with minimal input. The script authenticates via OAuth, then creates an event on your primary calendar using the Google Calendar API.
Demo Video
Watch the setup walkthrough on YouTube.
Prerequisites
- Python 3.11 (or use the provided Conda instructions).
- Google Cloud project with the Calendar API enabled.
- OAuth client credentials (
credentials.json) downloaded to this folder.
Recommended Conda Environment
conda create -n meeting_setup python=3.11
conda activate meeting_setup
pip install google-api-python-client google-auth-httplib2 google-auth-oauthlib mcp
Obtain credentials.json
- Visit the Google Cloud Console and sign in.
- Create/select a project.
- Enable Google Calendar API under APIs & Services → Library.
- Configure the OAuth consent screen (External is fine for testing) and add the Google accounts that will run this script as test users.
- Navigate to APIs & Services → Credentials and create an OAuth client ID of type Desktop app.
- Download the client configuration JSON, rename it to
credentials.json, and place it besidecreate_meeting_invite.py. - (First run only) When the script opens a browser window, approve access—
token.jsonwill be saved for subsequent runs.
Usage
python create_meeting_invite.py \
--meeting-time "2025-11-09 17:00" \
--other-email "karan.xyz@gmail.com" \
--meeting-title "Karan <> Arjun | Intro call" \
--meeting-notes "https://example.com/agenda"
--meeting-timeis required (IST timezone, formatYYYY-MM-DD HH:MMorYYYY-MM-DDTHH:MM).--other-emailis required.- Optional flags:
--meeting-title(default:Meeting)--host-email(default:arjuntheprogrammer@gmail.com)--meeting-notes(default: empty)--duration-minutes(default: 60)
The script prints the event details, including a link to view it in Google Calendar.
MCP Server
This project also exposes a Model Context Protocol (MCP) server so MCP-capable assistants can call the meeting tool directly.
conda activate meeting_setup
python mcp_server.py
The server currently runs over stdio transport. When connected, call the create_meeting_invite tool with the same argument names used by the CLI flags (e.g. meeting_time, other_email, meeting_title, meeting_notes, duration_minutes, host_email). The tool responds with key event details (summary, start, end, htmlLink, etc.).
Configure Codex to Use This MCP Server
- Ensure the MCP environment is active (step above) and Codex CLI is installed.
- Register the server once with Codex:
codex mcp add calendar_meeting /Users/arjungupta/anaconda3/envs/meeting_setup/bin/python /Users/arjungupta/Development/extra/google_calendar_meeting_setup/mcp_server.py - Verify the registration:
You should seecodex mcp listcalendar_meetingwith transportstdioand the Python command. - Launch Codex normally (
codex,codex exec, etc.). When Codex needs MCP tools it will start this server automatically. - Inside Codex, invoke the
create_meeting_invitetool by supplying the same arguments as the CLI flags, for example:create_meeting_invite meeting_time="2025-11-16 16:00" other_email="partner@example.com" meeting_title="Strategy Sync"
Remove or reconfigure the server any time with codex mcp remove calendar_meeting.
Configure Claude Desktop
- Launch Claude Desktop once so it generates
claude_desktop_config.jsonat~/Library/Application Support/Claude/. - Add (or merge) this entry inside the
"mcpServers"object:{ "calendar-meeting": { "command": "/Users/arjungupta/anaconda3/envs/meeting_setup/bin/python", "args": [ "/Users/arjungupta/Development/extra/google_calendar_meeting_setup/mcp_server.py" ] } } - Restart Claude Desktop (or toggle MCP servers in its settings) so it picks up the new server. The
create_meeting_invitetool is then available directly within Claude.
推荐服务器
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。
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。
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。
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。
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。
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。
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。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
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。