Everyday AI / Agentic Personal Assistant

MCP (Model Context Protocol) for an AI-powered personal assistant: Google Service MCP code is in this repo, plus config for Slack, Jira, and GitHub.

Overview

This repository contains:

• google-workspace-mcp/ — subfolder with the Google Service MCP (Gmail, Calendar, Meet, Docs). Build and run from there. See google-workspace-mcp/README.md for detailed features, tools, and troubleshooting.
• mcp-servers-config.json — Cursor MCP config (Google + Slack + Jira + GitHub)

Service	Description
Google Service	Gmail, Calendar, Meet, Google Docs integration
Slack	Slack messaging and channels
Atlassian (Jira)	Jira issue tracking
GitHub	GitHub repositories and issues

---

Setup

1. Google Service MCP

The Google Service MCP lives in the google-workspace-mcp subfolder of this repo. No separate clone — it’s part of this repo.

Build (from the subfolder)

cd google-workspace-mcp
npm install
npm run build

Configure Google Cloud Credentials

1. Go to Google Cloud Console
2. Create or select a project
3. Enable these APIs:
   • Google Meet API
   • Google Calendar API
   • Google Drive API
   • Gmail API
   • Google Docs API
4. Go to APIs & Services → Credentials
5. Click Create Credentials → OAuth client ID
6. Select Desktop app
7. Download and save as credentials.json inside the google-workspace-mcp folder (same folder as its package.json)

Authenticate

From the google-workspace-mcp folder:

cd google-workspace-mcp
echo '{"jsonrpc":"2.0","id":1,"method":"tools/call","params":{"name":"authenticate","arguments":{}}}' | node dist/index.js

A browser window will open for Google sign-in. After authorizing, token.json is created in google-workspace-mcp. Do not commit credentials.json or token.json.

Update Config Path

In mcp-servers-config.json (at repo root), point google-service at the google-workspace-mcp subfolder’s build:

"google-service": {
  "command": "node",
  "args": ["./google-workspace-mcp/dist/index.js"]
}

If Cursor’s workspace is this repo, ./google-workspace-mcp/dist/index.js is enough. Otherwise use the full path, e.g. /path/to/everyday-ai/google-workspace-mcp/dist/index.js.

---

2. Slack MCP Server

Uses the official Slack MCP Docker image.

Prerequisites

• Podman or Docker installed
• Slack Bot Token with appropriate permissions

Get Slack Credentials

1. Create a Slack App:
   • Visit the Slack Apps page
   • Click "Create New App"
   • Choose "From scratch"
   • Name your app and select your workspace
2. Configure Bot Token Scopes: Navigate to "OAuth & Permissions" and add these scopes:
   • channels:history - View messages and other content in public channels
   • channels:read - View basic channel information
   • chat:write - Send messages as the app
   • reactions:write - Add emoji reactions to messages
   • users:read - View users and their basic information
   • users.profile:read - View detailed profiles about users
3. Install App to Workspace:
   • Click "Install to Workspace" and authorize the app
   • Save the "Bot User OAuth Token" that starts with xoxb-
4. Get your Team ID (starts with a T) by following this guidance

Configure Environment

Set the following in mcp.json:

"slack-mcp-server": {
    "command": "podman",
    "args": [
    "run", "-i", "--rm",
    "-e",
    "SLACK_BOT_TOKEN",
    "-e",
    "SLACK_TEAM_ID",
    "-e",
    "SLACK_CHANNEL_IDS",
    "docker.io/mcp/slack"
    ],
    "env": {
    "SLACK_BOT_TOKEN": "<YOUR_SLACK_BOT_TOKEN>",
    "SLACK_TEAM_ID": "<YOUR_SLACK_TEAM_ID>",
    "SLACK_CHANNEL_IDS": "<YOUR_SLACK_CHANNEL_IDS>"
    }
}

---

3. Atlassian (Jira) MCP

Uses mcp-atlassian via uvx.

Prerequisites

• Python with uvx installed (pip install uvx)
• Jira Personal Access Token

Get Jira Token

1. Go to your Jira instance (e.g., https://issues.redhat.com)
2. Navigate to Profile → Personal Access Tokens
3. Create a new token with appropriate permissions

Configure

Update mcp-servers-config.json:

"mcp-atlassian": {
  "command": "uvx",
  "args": [
    "mcp-atlassian",
    "--jira-url",
    "https://your-jira-instance.com",
    "--jira-personal-token",
    "<YOUR_JIRA_PERSONAL_TOKEN>",
    "--transport",
    "stdio"
  ]
}

---

4. GitHub MCP

Uses GitHub Copilot's MCP endpoint.

Get GitHub Personal Access Token

1. Go to GitHub Settings → Developer settings → Personal access tokens
2. Generate a new token with required scopes (repo, read:org, etc.)

Configure

Update mcp-servers-config.json:

"github": {
  "url": "https://api.githubcopilot.com/mcp/",
  "headers": {
    "Authorization": "<YOUR_GITHUB_PAT>"
  }
}

---

Usage with Cursor IDE

1. Open Cursor IDE
2. Go to Settings → Features → MCP Servers
3. Add the contents of mcp-servers-config.json (with your actual secrets filled in)
4. Restart Cursor

Example Prompts

Once configured, you can use natural language to interact with these services:

Google:

• "Show my upcoming meetings"
• "Send an email to john@example.com about the project update"
• "Create a Google Doc called 'Meeting Notes'"

Slack:

• "Send a message to #general channel"
• "Check my Slack DMs"

Jira:

• "Show my open Jira issues"
• "Create a new bug ticket"

GitHub:

• "List my open pull requests"
• "Show issues in repo X"

---

Security Notes

• Never commit secrets to version control
• Store tokens securely (consider using environment variables or a secrets manager)
• Regularly rotate your tokens
• Use minimal required permissions for each service

---

File Structure

Google Service MCP is in the google-workspace-mcp subfolder. No git clone of a separate repo — everything is here.

everyday-ai/                          ← repo root
├── google-workspace-mcp/             ← Google MCP (Meet, Calendar, Gmail, Docs)
│   ├── src/
│   ├── dist/                         # Build output (npm run build from inside this folder)
│   │   └── index.js                  # ← Cursor runs this for google-service
│   ├── package.json
│   ├── credentials.json              # You add this (git-ignored)
│   └── token.json                    # Created after auth (git-ignored)
├── mcp-servers-config.json           # Cursor MCP config
└── README.md