AI Gateway Labs workflow
ai-gateway-labs
Jupyter labs on Azure API Management as a gateway for models, MCP tools and agents: token limits, tracing and OAuth. An Azure subscription is needed
- 1,003 ★ on the repository
- Last commit
- MIT
Broad access
Why this level
- Create paid Azure resources
- Contributor and RBAC Administrator rights are needed
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add workflows/ai-gateway-labsDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewBroad accessThe tool can change data, run commands or handle secrets.
- Create paid Azure resources
- Contributor and RBAC Administrator rights are needed
This is third-party code. Review the repository files before installing.
What it does
AI Gateway is a set of Jupyter notebook labs on using Azure API Management as a gateway for models, tools (MCP) and agents. The labs cover token limits, tracing and OAuth. Installation: `git clone https://github.com/Azure-Samples/AI-Gateway.git` and `uv pip install -r pyproject.toml`. An Azure subscription with Contributor and RBAC Administrator rights is needed, and resources are paid. MIT license, the repository was updated on 16.09.2026.
Who it is for. For platform engineers and Azure developers who learn to build a gateway for models and agents on API Management.
Good fit when
- You need to study token limits and tracing on a gateway
- You need OAuth examples for MCP tools
- You need API Management labs for agents
Not a fit when
- You have no Azure subscription: the labs create paid resources
- You need a ready gateway: these are learning samples
Example request
Go through the token limit lab on the gateway and show what the client receives when the limit is exceededLimitations
The labs create resources in your Azure subscription and need Contributor and RBAC Administrator rights, so access is wide and resources are paid. Use a separate test subscription and delete resources after work. These are learning samples and not a ready solution.
How to disable. Delete the Azure resources created by the labs and the repository clone.
FAQ
What is it?
Jupyter labs on Azure API Management as a gateway for models, MCP and agents.
What is needed to work?
An Azure subscription with Contributor and RBAC Administrator rights, with paid resources.
Is it a ready gateway?
No, these are learning samples.
Discussion
No questions or reviews yet.
Related
WorkflowsMCP Gateway
mcp-gateway
A reverse proxy and lifecycle management for MCP servers on Kubernetes and Azure from Microsoft: routing, authorization, roles. For platform teams
Agentgateway
agentgateway
A proxy for agentic traffic: routes and secures connections to MCP servers, LLMs, and other agents over A2A
Azure Skills
Azure Skills Plugin
Microsoft's official plugin: Azure skills plus the Azure MCP Server and Foundry MCP wired in for real cloud actions
Microsoft Learn skills for nearly every Azure service: limits, configuration, security, architecture and troubleshooting with doc links
Microsoft's official MCP server for Azure: the agent sees subscription resources and works with Azure services on your behalf
Azure DevOps MCP server
Azure DevOps MCP Server
Official Microsoft MCP server for Azure DevOps: work items, repositories, builds, test plans, and wiki from the agent
README in short
The README describes Jupyter labs on using Azure API Management as a gateway for models, MCP tools and agents. MIT license.