Are these steps for the ordinary Microsoft Copilot app?

No. They add a tool to an agent in Copilot Studio. GitHub Copilot also has its own MCP setup, separate from these instructions.

Can I test this with a Copilot Studio trial?

A trial can be used to test an agent, but it cannot publish the agent. The limited Teams plan does not include these features.

Which authentication option should I choose?

Choose OAuth 2.0, then Dynamic discovery. Create a new connection and complete the fenbs sign-in and approval before adding it to the agent.

Does Copilot Studio still support SSE MCP servers?

Microsoft says SSE is no longer supported after August 2025. The documented connection uses Streamable transport.

Connect Copilot Studio to your fenbs kanban board

Copilot Studio makers need a standalone plan, or a trial for testing, and permission to add tools. Connect fenbs to your agent, then save decisions and next steps for the next AI to read.

Last checked October 28, 2026

You need Copilot Studio maker access on a standalone plan, or a trial for testing, with generative orchestration turned on. Your organization can limit connections. The limited Teams plan does not include these features, and a trial can test an agent but cannot publish it. Check Microsoft’s MCP requirements (opens in a new tab) and plans and trials (opens in a new tab).

Connect Copilot Studio to fenbs

Open your agent in Copilot Studio. Choose Tools, then Add a tool, then New tool.
Choose Model Context Protocol. Enter fenbs as Server name, add a short Server description, and paste the address below into Server URL.
Choose OAuth 2.0, then Dynamic discovery. Choose Create, then Next.
Choose Create a new connection. Finish the fenbs sign-in and approval, then choose Add to agent.
Ask your agent: “What is on my fenbs kanban board?”

The server URL

https://fenbs.ai/api/mcp

Microsoft’s existing MCP server guide (opens in a new tab) applies to agents and flows using its standard harness. It supports Streamable transport; the old SSE transport is no longer supported. These are Copilot Studio settings, separate from the ordinary Microsoft Copilot app and from GitHub Copilot. Set it up in fenbs for the short setup steps.

What to save from a Copilot Studio session, and how the next AI picks it up

After working with your agent, ask it to save the useful result in fenbs. Record what you decided, the reason for the choice, facts you checked and the next step. If something still needs a person’s answer, say exactly what is waiting. Keep the record focused on what someone needs to continue the work.

Save agreed choices with their reasons, so they are not reopened by accident.
Record what was tried, what worked and what still needs checking.
Leave unfinished tasks with a clear next step.

Next time, ask the connected AI to read the saved work in fenbs first. It can be this Copilot Studio agent or a different connected assistant. The saved decisions and next steps give it a place to begin without relying on the previous conversation being in its chat history.

How fenbs helps

Never lose the work you did in an AI session

After a useful AI session, save it in fenbs: the key observations, findings, research and context. Then carry on from exactly where you left off, on any device and in any other AI. The next AI you connect reads it at the start of the session, so you never have to explain the same problem again.

Create a free account

Free for up to 5 people and 50 open tasks · No credit card · Works with Claude, ChatGPT, Cursor and more