⚠️ Important note
The example graph and key for this workshop is now unavailable so you will not be able to follow along. This workshop companion is still available for reference purposes.
This workshop companion was designed to be used alongside an instructor for GraphQL Summit 2025. If you would like to learn on your own time at your own pace, check out the Odyssey course instead: Agentic GraphQL: MCP for the Enterprise
This workshop uses Apollo MCP Server v1.0.0. If you are using later versions, the instructions may differ slightly. For example, v1.1.0 uses port 8000 as the default port instead of 5000.
Overview
Agentic AI unlocks powerful new capabilities, but with that power comes new risks. Without the right safeguards, these systems can expose sensitive data, execute unvetted logic, or become unpredictable. In this workshop, you'll take an existing MCP Server and learn best practices on how to secure it. You'll use authentication, persisted queries, introspection, contracts, and more to give your AI assistants the structure and guardrails they need to be helpful, without going rogue and DDOS-ing your systems.
What you'll do
- Set up authentication with the Apollo MCP Server
- Explore the different ways of restricting access to your data with persisted queries, introspection and contracts using Apollo GraphOS
Who this workshop is for
Anyone looking to implement AI agents in their organization.
Pre-requisites
Basic experience with the Apollo MCP Server
Let's get set up!
Clone the repository
In your terminal, run the following command to clone the repository.
git clone https://github.com/apollographql-education/summit25-workshop-mcp-security.git
Install the Rover CLI
You need at least Rover version v0.35.0 to complete the workshop.
Open up a terminal and run the install command that suits your computer's environment:
For Linux / Mac OS:
curl -sSL https://rover.apollo.dev/nix/latest | shFor Windows PowerShell installer
iwr 'https://rover.apollo.dev/win/latest' | iexNote: You can find other installation methods in the Apollo documentation.
Verify that the installation completed successfully by running
roveranywhere in the terminal.If it outputs a list of options and subcommands for using Rover, great! The CLI is installed and ready to go.
Download Claude Desktop
We'll be using Claude Desktop as our AI agent or LLM for this workshop.
Download and install it from Claude's official website.