A number of weeks in the past at BUILD, we introduced the upcoming Azure DevOps MCP Server:
👉 Azure DevOps with GitHub Repositories – Your path to Agentic AI
As we speak, we’re excited to share that the native Azure DevOps MCP Server is now obtainable in public preview. This lets GitHub Copilot in Visible Studio and Visible Studio Code entry and work together together with your Azure DevOps surroundings, together with work objects, pull requests, check plans, builds, releases, and wiki pages.
🤷♂️ What’s an MCP Server?
An area MCP Server (Mannequin Context Supplier) is a software that sits between your AI assistant (like GitHub Copilot) and your Azure DevOps group. Its job is to inject wealthy, real-time context resembling work objects, pull requests, check plans, and so forth., into the prompts despatched to the LLM. This lets the assistant give higher, extra correct, and extra related solutions tailor-made to your particular Azure DevOps venture.
Not like cloud-based options, an area MCP Server runs inside your community or growth surroundings. Which means it could safely entry personal knowledge and instruments with out that knowledge ever leaving your system.
📽️ Instance
You’ll find a number of instance movies and prompts within the how-to information. One enjoyable instance reveals how one can take a person story, generate a set of check circumstances with detailed steps based mostly on the work merchandise description, after which hyperlink these check circumstances again to the unique person story.
If in case you have concepts or want examples for particular eventualities, be at liberty to create a problem within the repository.
⭐ Getting Began
Getting began is comparatively simple, however it does require a number of setup steps. Here’s a fast overview of the steps, however we do advocate you observe the total set up and setup directions offered within the repository to make sure every part is configured accurately.
- Navigate to the Azure DevOps MCP Server repository.
- Check in utilizing the Azure CLI:
az login
- Copy and paste the configuration into your native
.vscodemcp.json
file. - Begin the MCP Server and start utilizing the instruments.
💬 Suggestions
We’re actively accumulating suggestions all through the general public preview stage. In the event you run into any points or have ideas, we’d love to listen to from you. Please create or touch upon Points immediately within the Azure DevOps MCP Server repository.