AI Time Tracking with Teetrack: The Complete Guide (MCP, Claude, ChatGPT, Calendar)
"Log two hours on the website project" should be all it takes to track time. With Teetrack's MCP server and calendar automation, it is. This guide is the map to everything AI-related in Teetrack: what the MCP server does, how to connect each assistant, and how calendar auto-import removes the last bits of manual entry.
What this guide covers:
- What AI time tracking actually means (and what it doesn't)
- Setting up the MCP server with Claude, ChatGPT, or GitHub Copilot
- Auto-importing calendar events as time entries
- Real workflows that combine both
What "AI time tracking" means in Teetrack
Teetrack does not guess what you worked on by watching your screen — it never takes screenshots or monitors activity. Instead, it gives AI assistants a safe, structured way to operate your time tracking for you. You stay in control of what gets recorded; the assistant just removes the typing.
Two building blocks make that possible: the MCP server, which lets assistants start timers, log time, and query your history, and calendar integration, which turns meetings you already scheduled into draft time entries automatically.
The MCP server: your time tracker, as a set of AI tools
The Model Context Protocol (MCP) is the open standard AI assistants use to call external tools. Teetrack ships a native MCP server that exposes your account as tools an assistant can use: start and stop timers, log completed time, list projects and clients, pull time summaries, and check budget status. If you can phrase it, the assistant can usually do it — "what did I work on last week?", "start a timer on Client X", "log yesterday's 3 hours of design work".
Start with the announcement post, Introducing the Teetrack MCP server, then follow the setup guide for your assistant:
- Claude — setup guide, plus a hands-on walkthrough of daily tracking with Claude
- ChatGPT — setup guide; the ChatGPT & Gemini post covers both ecosystems
- GitHub Copilot — track development time without leaving your editor
Once the basics work, the advanced MCP workflows post shows what becomes possible when you chain tools: weekly reviews written by your assistant, budget checks before accepting new work, and automated timesheet prep.
Calendar automation: the other half of hands-off tracking
Most billable time is already written down — in your calendar. Teetrack syncs Google Calendar, Outlook, and any iCal feed, and can auto-import events as draft time entries. With smart project matching, recurring client meetings land in the right project automatically, based on keywords in the event title.
Not sure which calendar source fits your setup? The Google Calendar vs Outlook vs iCal comparison breaks down the differences.
Putting it together: AI + calendar in one workflow
The combination is where manual time entry effectively disappears: the calendar integration drafts entries for your scheduled meetings, and your assistant fills the gaps — logging the unscheduled deep work, reassigning mismatched entries, and summarizing the week when it's time to bill. The AI-managed calendar tracking post walks through a full week run this way.
A realistic daily loop looks like this:
- Morning: calendar sync has already drafted entries for today's meetings.
- During the day: "start a timer on the API project" — spoken to your assistant, not typed into a form.
- End of day: "log 90 minutes of code review on Client Y, billable."
- Friday: "summarize my billable hours this week by project" — and the numbers are ready for your invoice.
Where to go next
If you're new to Teetrack, create a free account first — the free tier includes the MCP server and calendar sync. Then connect your assistant via the MCP guide hub and your calendar via the calendar guide hub. The whole setup takes about ten minutes, and it's the last time tracking chore you'll do by hand.