DevProxy version 0.29 introduces AI capabilities to help configure the tool using natural language. It simulates various API behaviors, including rate-limiting and throttling, aiding developers in resilience testing. A new MCP server enables AI agents, such as GitHub Copilot, to automate configurations. Internal usage of language models allows DevProxy to generate OpenAPI and TypeSpec files from API call samples. The update includes required adjustments for plugin development due to breaking changes in architecture. Enhanced troubleshooting features help developers quickly identify issues. Automatic uninstallation of root certificates adds to user convenience.
DevProxy (formerly known as Microsoft 365 Developer Proxy) helps developers to add resilience to their API-related code by simulating a vast number of API and network behaviours.
The new version brings breaking changes that will require some adjustments for developers building DevProxy plugins. The new architecture uses a different base class coming from DevProxy.Abstractions assembly.
DevProxy internally uses language models (LLMs) to generate OpenAPI or TypeSpec files from API call samples, enhancing its capabilities in API simulation.
To support DevProxy configuration using natural language, the development team has released a MCP server. It allows AI agents to automatically configure DevProxy for user's needs.
Collection
[
|
...
]