Does custom MCP require a paid Replit plan?

The custom-MCP setup guide does not state a paid-plan requirement. Available integrations can depend on your workspace plan, so check the options in your workspace.

Is this the Replit MCP Server for other assistants?

No. These steps connect Replit Agent to fenbs as an external tool. Replit also offers its own MCP Server for other clients to work with Replit Apps; that is a separate connection direction.

Can I give Replit a local stdio command?

The setup documented here uses a remote HTTPS server. The sources do not establish support for a local stdio process on your laptop.

Does saving progress in fenbs replace Replit checkpoints?

No. Save the conclusions and next steps in fenbs for the next connected assistant. Use Replit checkpoints to restore Replit state; a saved summary is not a backup of the app or its data.

Connect Replit to your fenbs kanban board

Add fenbs to Replit Agent as a custom MCP server. Sign in through fenbs, then keep the useful results of a build session where your next connected AI assistant can read them.

Last checked October 28, 2026

Connect Replit to fenbs

Replit Agent can connect to external tools through remote MCP servers. The setup docs do not name a required plan or country for a custom server; available integrations can depend on your workspace plan. These steps are for the web editor and settings.

In your Replit Workspace, open Integrations, then the MCP Servers settings pane. Choose + Add MCP server.
Set Display name to fenbs and paste the URL below into MCP Server URL.
Click Test & save. Follow the fenbs sign-in page and approve the connection.
Return to Replit Agent and ask: “What is on my fenbs kanban board?”

The MCP server URL

https://fenbs.ai/api/mcp

Replit’s Connect via MCP guide (opens in a new tab) describes custom HTTPS servers, OAuth sign-in and optional headers. The connection makes tools available across your Replit work. Some tools require confirmation, and Replit’s security scanner can reject tools. Set it up in fenbs.

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

Before ending a build session, ask Replit Agent to save the useful conclusions in fenbs: what changed, what you checked, what still fails and the next step. A short account of a failed approach is useful too, so the next assistant has a reason to try something else.

The task you were working on and the files that changed.
The result of a test or preview, including anything you did not check.
A decision you made and why it matters to the next change.
The next action and any question waiting for you.

At the start of the next session, ask the connected AI to read fenbs before making changes. It can pick up the saved work whether you return to Replit or use another connected assistant. Replit’s replit.md file (opens in a new tab) remains useful for instructions inside Replit, but those instructions do not automatically apply to other AI tools.

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

What this connection does not move

Keep Replit’s checkpoints for restoring its working state. Saving a summary in fenbs does not move a checkpoint, a running app or a database. Replit’s rollback guide (opens in a new tab) says development database restoration is optional and a published app needs publishing again after a rollback.