Model Context Protocol

Securely Deploy AI & Agentic Workflows Across Your Organization

The Artist Growth MCP server connects Claude, ChatGPT, Copilot, and Gemini directly to your live roster — schedules, events, tasks, documents and reporting — with your existing permissions enforced on every request.

Start for free See how it works

What is MCP?

A standard way for AI to use your tools

The Model Context Protocol is an open standard that lets an AI assistant talk to software the same way a person would — asking questions, pulling records, filing requests. Instead of copy-pasting itineraries into a chat window, you point your assistant at Artist Growth once and it works with the real thing.

Nothing about your account changes. The assistant sees exactly what the person using it is allowed to see, down to the field level, and every write it makes is logged like any other action in the platform.

  • Open standard

    Not a bespoke plugin. One connection, any compliant client.

  • Live data

    Every answer reads the current state of your account.

  • Your permissions

    User- and field-level access carries straight through.

Works with the assistant your team already uses

Plus any MCP-compatible client

  • Claude
  • ChatGPT
  • Microsoft Copilot
  • Gemini

Try asking

Prompts that replace an afternoon

Every answer comes from live Artist Growth data — not a stale export.

  • Get me the run of show for Northern Wild’s July 31 show at Beacon Theatre.
  • Which shows next month are missing hospitality riders?
  • Build a report of routing gaps over 400 miles for the fall run.
  • File four guest ticket requests for the Nashville date under my name.
  • Draft a day sheet for tomorrow and flag anything still unconfirmed.
  • Who on my team has more than ten open tasks this week?

Capabilities

Six tools, permissions intact

We’re adding more all the time. New tools show up in your assistant automatically — nothing to reinstall.

  • list_calendars

    Read

    Whole-schedule overview across one or more artists.

    Each calendar consolidates everything in a date window — events, day-of logistics (load-in, soundcheck), transportation, and accommodations. Window required, max one year.

  • search_events

    Read

    Finds events by free text, date range, artist, status, and/or event type.

    Use list_calendars instead when you want the full schedule picture rather than filtered events.

  • list_artists

    Read

    Resolves a specific artist to their identifier, or answers “which artists can I access?”

    Not a required first step — the other tools already span every accessible artist by default.

  • list_my_tasks

    Read

    The dedicated answer to “what’s mine / what’s overdue / what’s due.”

    Scoped to the signed-in caller. Narrows by role (assignee, creator, assigner), org, project, event, due-date window, or priority. Incomplete-only by default; each task carries a derived due status.

  • search_ticket_requests

    Read

    Finds ticket requests — guest, submitter, approver, department, requested tiers and passes, and the event.

    Filters by event, artist, department, event group, status, event-date window, and submission-date window.

  • create_ticket_request

    Write

    Opens an interactive form to request tickets and passes.

    Pick an event by search or pass one in; tiers, passes, and departments populate from that event. Guest details can be prefilled. The user completes and submits — this isn’t a write the model drives on its own.

Connected in less than 2 minutes

No engineering ticket required.

  1. Enable the server

    Your chosen LLM’s admin enables the Artist Growth MCP and optionally sets tool restrictions.

  2. Connect your client

    Each user adds the Artist Growth connection in Claude, ChatGPT, Copilot or Gemini and signs in once.

  3. Start asking

    The assistant discovers the available tools automatically. No prompt engineering needed.

Security & permissions

Force AI into custom permissions, not generic roles

Access is scoped to the authenticated user. If a tour manager can’t see a settlement in the platform, their assistant can’t either.

  • Scoped to the signed-in user Requests authenticate as a person, never as the organization. It’s API-backed — the same trusted API you’ve used for over a decade, called on the current user’s behalf.
  • Field-level enforcement Hidden fields stay hidden — including in generated reports and documents.
  • Full audit trail Every read and write is logged with the user, client and timestamp.
  • Writes need a human Write tools open a form for the user to review and submit — the model never commits a change on its own.
  • Prompt injection mitigation Proprietary safeguards inspect every tool call, so instructions hidden in your data can’t redirect the assistant.
  • No training on your data* Artist Growth never sends your records to model providers for training. What your AI provider does is set in your workspace.

* Artist Growth never sends your records to a model provider for training. Beyond that, training is controlled by you, in your AI provider’s settings — and on business plans it is off by default: Anthropic does not train on inputs or outputs from its commercial products, and OpenAI excludes ChatGPT Business workspace data by default. Consumer plans may differ, but these options are under your control.

Questions

Do I need a developer to set this up?

No. An admin enables the MCP server in account settings and each user pastes a connection URL into their AI client. Nothing is installed on your machines.

Can the AI change data in my account?

Only what the signed-in user could change themselves — creating events, filing ticket requests, assigning tasks. Writes are attributed to that user and appear in the activity log.

Which assistants are supported?

Any client that speaks MCP. Today that includes Claude, ChatGPT, GitHub Copilot and Gemini, and the list grows as clients adopt the standard.

Does connecting bypass our SSO or MFA?

No. Users authenticate through the same sign-in flow as the web app, including SSO and MFA where your account requires it.

Does it work on mobile?

Absolutely it does. If your LLM has a mobile client, it works out-of-the-box.

Is my roster data used to train models?

Artist Growth never sends your records to a model provider for training. Beyond that, training is controlled by your AI provider settings — and on business plans it is off by default. Anthropic does not train on inputs or outputs from its commercial products, including Claude for Work and the API, and OpenAI excludes ChatGPT Business and Enterprise workspace data from training by default. Consumer plans may differ, but these options are under your control.

What does it cost?

The MCP server is included with every Artist Growth plan at no additional cost.

Put your roster operations in front of your AI

Free to start. The MCP server is included with every Artist Growth account.

Start for free Book a demo