Connect an AI assistant (MCP)
Linyup runs a remote MCP (Model Context Protocol) server, so you can connect an AI assistant to your studio and ask questions in plain language:
- “Who hasn’t come in three weeks?”
- “How full were Thursday classes last month?”
- “What did we earn from drop-ins in August?”
Access is read-only, and limited to the scopes you grant.
What the assistant can look up
Section titled “What the assistant can look up”| Tool | What it answers |
|---|---|
get_studio_overview |
A summary of your studio |
find_contacts, get_contact, get_contact_history |
Finding people and their history |
list_inactive_contacts |
Who has stopped coming |
get_schedule, get_session_roster, list_events |
What’s on, and who’s booked |
list_offerings, list_memberships |
What you offer and who holds which plan |
get_attendance_trend, get_class_fill_rates |
Attendance and fill rates over time |
get_revenue_summary |
Revenue |
Connecting
Section titled “Connecting”The server address is https://api.linyup.com/mcp. It needs the API & AI connectors add-on.
claude.ai, ChatGPT, VS Code and other assistants with remote connectors: add a custom connector with that address. The assistant sends you to Linyup to sign in, shows the studio and the scopes it asks for, and connects once you approve. You can see and disconnect connected apps under Settings → API keys.
Claude Code and other tools that take a header: create an API key under Settings → API keys (the screen shows this command with your key filled in), then run:
claude mcp add --transport http linyup https://api.linyup.com/mcp \ --header "Authorization: Bearer lyp_live_…"The assistant sees only what the key’s scopes allow, and only what you yourself may see in the app.