Your Lifehub
inside your AI chat
Lifehub's MCP exposes over 90 tools that any compatible AI assistant can use to read, create, and update your accounts, expenses, goals, habits, tasks, and financial analytics. Set it up in 2 minutes.
What is MCP?
Model Context Protocol is an open standard created by Anthropic to connect applications to language models. It's like a USB for AI assistants.
How does it work?
You add https://applifehub.com/mcp to your favorite AI client, sign in to Lifehub in the browser window that opens, and pick which areas it may access. Done — no keys to copy.
Is it safe?
Authentication uses OAuth 2.1 with PKCE. Access tokens are short-lived (2 hours), refresh automatically, are hashed at rest, and limited to 60 requests per minute. Every call is logged.
What you can do
Your AI assistant,
now handling your routine
Talk to the AI the way you'd talk to a personal advisor — it accesses your Lifehub and gets things done in seconds.
Finance at your fingertips
Accounts · Expenses · Debts · Investments · Goals
Track balances, log expenses by voice, run projections and decisions — without opening the platform.
- ›How much did I spend on food this month?
- ›Log an expense: Uber $5, today, transportation.
- ›If I keep saving at this pace, when do I hit my emergency fund goal?
Habits, workouts, and routine
Habits · Sports · Birthdays · Countdowns
Check off habits hands-free, log workouts from your phone, never miss an important birthday again.
- ›Mark my reading habit as done today.
- ›Log a gym workout, 60 minutes, intensity 4.
- ›What birthdays do I have in the next 30 days?
Personal organization
Notes · Lists · Kanban · Construction
Create kanban cards, add items to your grocery list, take quick notes — wherever you're chatting from.
- ›Add 'bread, milk, eggs' to the grocery list.
- ›Create a kanban card: 'Review budget', high priority.
- ›How much have I spent on the kitchen renovation so far?
Insights and progress
Gamification · Analytics · Charts
Get analysis in natural language, generate charts on the fly, see your gamified progress without opening dashboards.
- ›Show my net worth month by month over the last 12 months.
- ›What's my current level and how much XP to the next one?
- ›Compare my spending with last month by category.
For developers and advanced users
Setup
Connect in 2 steps
Any MCP client that speaks Streamable HTTP only needs the server URL — sign-in happens in your browser via OAuth.
Add the server
Add the Lifehub MCP server to your client: https://applifehub.com/mcp. No API keys or tokens to copy.
Sign in and authorize
Your client opens a browser window — sign in to Lifehub and choose which areas (finances, habits, notes, ...) the client may access. Access tokens are short-lived and refresh automatically.
Sign in to LifehubClaude Desktop
Anthropic's official app for macOS and Windows
- Open Settings → Developer → Edit Config (or edit
claude_desktop_config.jsonmanually). - Add the server below.
- Restart Claude Desktop and approve the connection — a browser window opens so you can sign in to Lifehub and choose what it may access. The tools icon 🔧 should then list all Lifehub tools.
{
"mcpServers": {
"lifehub": {
"type": "http",
"url": "https://applifehub.com/mcp"
}
}
}
💡 Config path: ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows).
Claude Code
Anthropic's CLI for development
Add it via command:
claude mcp add --transport http lifehub https://applifehub.com/mcp
Or add it manually to ~/.claude/mcp.json:
{
"mcpServers": {
"lifehub": {
"type": "http",
"url": "https://applifehub.com/mcp"
}
}
}
💡 Run /mcp inside Claude Code to authenticate — your browser opens for the Lifehub sign-in.
ChatGPT
OpenAI's app and web (via Connectors)
- Go to Settings → Connectors in ChatGPT (web or app).
- Click Add custom connector.
- Fill in:
- • Name: Lifehub
- • URL:
https://applifehub.com/mcp - • Authentication: OAuth (ChatGPT opens the Lifehub sign-in in your browser)
- Enable the connector in the tool selector during your conversation.
💡 Connectors are available on Plus, Pro, Business, Enterprise, and Edu plans. Requires enabling "Developer mode" in some cases.
Gemini CLI
Google Gemini CLI and Gemini Code Assist
Edit ~/.gemini/settings.json:
{
"mcpServers": {
"lifehub": {
"httpUrl": "https://applifehub.com/mcp"
}
}
}
💡 Verify with /mcp list inside Gemini CLI.
Cursor
AI-powered code editor
Create or edit ~/.cursor/mcp.json (or .cursor/mcp.json in your project):
{
"mcpServers": {
"lifehub": {
"url": "https://applifehub.com/mcp"
}
}
}
💡 Go to Cursor Settings → MCP and enable the lifehub server.
Windsurf
Codeium's AI editor
Open Windsurf Settings → Cascade → MCP Servers → Add server, or edit ~/.codeium/windsurf/mcp_config.json:
{
"mcpServers": {
"lifehub": {
"serverUrl": "https://applifehub.com/mcp"
}
}
}
Other MCP clients
Zed, Continue, any compatible client
Lifehub uses Streamable HTTP with OAuth 2.1 (authorization code + PKCE, dynamic client registration). If your client supports MCP authorization, just pass the URL and complete the sign-in it opens:
https://applifehub.com/mcp
Check the endpoint via curl — unauthenticated requests get 401 with a WWW-Authenticate header pointing at the OAuth metadata:
curl -i -X POST https://applifehub.com/mcp \
-H "Content-Type: application/json" \
-H "Accept: application/json, text/event-stream" \
-d '{"jsonrpc":"2.0","method":"tools/list","params":{},"id":1}'
Examples
What to ask the AI?
A few prompts to start — the AI will pick the right tool on its own.
"Analyze my spending from the last 3 months and identify the 3 categories where I spend the most."
"Log 3 expenses: lunch $9, Uber $5, movie $12, all today."
"If I keep saving at this rate, when do I hit my $20k goal?"
"Which habits did I keep the longest streak on this year?"
"Create a card on my kanban board: 'Review budget' in the This Week column, high priority."
"Generate a bar chart of my net worth month by month over the last year."
Security
You're in control
OAuth 2.1 with PKCE
Authorization-code flow with PKCE — no long-lived secrets in URLs or config files. Your client signs in through the browser, just like a trusted app.
Short-lived tokens, hashed at rest
Access tokens expire after 2 hours and rotate via refresh tokens. All secrets are stored hashed — never in plain text.
Per-client permissions
On the consent screen you pick which life domains (finances, habits, notes, ...) each client can access. Want analytics read-only? Grant just that.
Instant disconnect
One click in Settings → MCP disconnects a client immediately. No subsequent call goes through.
Full activity log, isolated per user
Every call made by the AI is logged at /settings/mcp/activity: tool, arguments (with sensitive keys masked), date, and response code. Each client is bound to your user — accessing another person's data is impossible by architecture.
Rate limiting
60 requests per minute per client — plenty for normal use, protects against misbehaving AI loops.
FAQ
MCP questions, answered
How AI control of your Lifehub data works.
What is MCP (Model Context Protocol)?
Which AI clients work with Lifehub MCP?
How do I set up the Lifehub MCP server?
Is MCP access to my data secure?
What can the AI actually do with my Lifehub data?
Ready to connect?
Create your account, connect your AI client with one sign-in, and chat with your financial life from the AI you already use.