Basecamp MCP: How to Connect Basecamp to AI in 2026
No — Basecamp does not have an official MCP server. That’s not for lack of interest in AI: in March 2026, 37signals shipped an “agent-accessible” Basecamp, but they did it with a revamped API, a brand-new basecamp-cli, and an agent skill — deliberately not a Model Context Protocol server. So what you’ll actually find searching “Basecamp MCP” are community-built servers, each wiring its own OAuth app into the Basecamp API — not an MCP integration Basecamp built or endorses.
Even where one of those community servers works well, the same limit applies: an MCP server hands Basecamp to an AI inside a conversation you start. It’s a doorway, not a worker. Nothing watches a project for you, nothing fires when a to-do gets assigned or a message lands on a board, and nothing runs while the chat is closed. Here’s the honest state of Basecamp and MCP, what the community servers actually do, where the limits are — and what to use when you want Basecamp work that runs on its own.
What an MCP connection to Basecamp does
Model Context Protocol (MCP) is the open standard that lets an AI client — Claude, ChatGPT, Cursor, and others — talk to an outside app through a shared interface. Basecamp hasn’t built one; it chose a CLI-plus-skill path instead. The servers people mean when they say “Basecamp MCP” are independent projects — georgeantonopoulos/Basecamp-MCP-Server, jhliberty/basecamp-mcp-server, stefanoverna/basecamp-mcp, and vapvarun/basecamp-mcp-server, among several others — each calling the Basecamp API under the hood through a developer app you (or the tool’s maintainer) register.
With one connected, an AI client can typically:
- Read projects and their contents — pull a project’s to-do lists, message board posts, docs, and schedule into a chat.
- Work with to-dos — list open to-dos, create new ones, check items off, and reassign them across a project.
- Post and comment — add a message to a board, drop a comment on a to-do, or write in a card table, using your Basecamp account’s own OAuth credentials.
- Search and summarize — look across a project or account for a keyword and have the AI summarize what’s happening.
It’s genuinely useful for ad-hoc work — “summarize where the launch project stands and list every to-do with no assignee” — answered from live Basecamp data instead of a stale guess.
How to connect Basecamp to AI in 2026
There’s no “install the official server” step, because Basecamp built a CLI and skill instead of an MCP server. The realistic MCP path is:
- Register a Basecamp integration. Go to launchpad.37signals.com/integrations, create an OAuth application, and get a client ID and secret. Essentially every community Basecamp MCP server needs this — you’re authenticating as your own developer app, not through Basecamp’s blessing of the MCP project itself.
- Pick a community server and check what it covers. Tool counts vary a lot between projects — some expose a few dozen tools, others claim 75 or more across projects, to-dos, message boards, campfires, card tables, docs, and search. Read the repo before trusting it with write access.
- Add the server to your AI client, point it at your Basecamp OAuth credentials, and restart the client so the tools register.
- Test with a read query first — pull a project’s to-do lists — before trusting any server with posting, commenting, or completing tools, since those act on your account and are visible to your whole team.
For most people searching “connect Basecamp to AI,” this is a real setup with real account access behind it, not a one-click connector. (If you’re a developer, Basecamp’s own CLI and skill are the officially supported route — but that’s a terminal tool for agentic coding, not a chat connector.)
Where the Basecamp MCP stops
Even setting the “which community server do I trust” question aside, the same four limits show up here:
- It only works inside a chat you start. Close the window and nothing happens. Nothing is watching a project or a message board for you.
- No triggers. A to-do getting assigned to you, a new message on a board, a deadline arriving, a comment mentioning your name — none of these can start anything through MCP. There’s no “when this happens in Basecamp, do that.”
- It’s one app at a time. A Basecamp MCP server knows Basecamp. Getting a completed to-do into a time-tracking sheet, a Slack channel, and an invoice means wiring up a separate MCP server for each and hoping your client can juggle them in one turn.
- You own the plumbing and the account risk. There’s no vendor support line here — you registered the OAuth app, and a server with write tools acts as your account, posting and completing things your whole team can see, with no one else reviewing what it does.
So a Basecamp MCP server, community-built as all of them are, is not a way to make Basecamp run — to have work happen on a schedule or in reaction to an event, across the other tools a project touches.
Running Basecamp work that doesn’t need a chat open
That “run on its own, across apps” gap is exactly where Carly fits. Carly connects to Basecamp and to the ~260 other apps it supports natively, plus anything with a public API through your own key — no community MCP server to vet, no separate developer app to register and maintain yourself. The difference from MCP is the important part: Carly’s workflows are triggered and scheduled, so Basecamp work happens whether or not anyone has a chat window open.
A few things a Carly workflow can do that no MCP server can:
- When a to-do is assigned to you in Basecamp → check it against your calendar, and if the due date is tight, draft a heads-up message and add a reminder.
- Every morning → gather to-dos due today across your projects and send a single digest to Slack.
- When a to-do is marked complete → log the hours to Harvest and add a line to the client’s invoice in QuickBooks.
The non-AI steps — the moving, matching, and routing between apps — are free and unlimited, the Zapier-style backbone of the workflow. The AI steps (drafting, summarizing, deciding) start at $35/month. You describe the outcome in plain language and Carly wires up the Basecamp connection and everything downstream.
If you just want to ask questions about a project from a chat, a community MCP server is the closest thing available, with a Basecamp OAuth app and its API terms behind it. If you want Basecamp activity to actually trigger something — on a schedule, in reaction to an event, across every app a to-do or message touches — that’s the job no Basecamp MCP server was built for.
FAQ
Does Basecamp have an official MCP server?
No. In March 2026, 37signals made Basecamp “agent-accessible” through a revamped API, a basecamp-cli, and an agent skill — but deliberately not an MCP server. What exists for MCP are community projects — such as georgeantonopoulos/Basecamp-MCP-Server, jhliberty/basecamp-mcp-server, and stefanoverna/basecamp-mcp — that call the Basecamp API through an OAuth app you register yourself.
Is there a free way to connect Basecamp to an AI over MCP? The community servers are open source and free to run, and a Basecamp OAuth app is free to create at launchpad.37signals.com/integrations. You still need a Basecamp account, and you’re responsible for hosting and updating whichever server you choose.
Can a Basecamp MCP server trigger automations? No. MCP is request/response inside an AI chat — it has no triggers and nothing runs when the conversation is closed. For event- or schedule-driven Basecamp work across apps, you need a workflow tool like Carly rather than an MCP server.
Can I connect Basecamp to AI without coding or hosting a server? Yes. You don’t have to register a Basecamp OAuth app or run an MCP server at all. Carly connects to Basecamp for you and lets you build the automation in plain language — describe what you want to happen and it wires up the Basecamp connection and the other apps involved, with no server to host and no code to write.
Ready to automate your busywork?
Carly schedules, researches, and briefs you—so you can focus on what matters.
See what people say
"Before Carly, I relied on a Calendly link, but the whole process felt impersonal and not very professional. Carly changed that by handling all the back-and-forth, so I'm no longer stuck in endless email threads trying to line up schedules.
Now Carly reaches out to candidates, shares my real-time availability, lets them pick a slot, then sends a Zoom link and drops it straight into my calendar. She sends reminders to both of us before each call, which has significantly reduced no-shows and last-minute confusion.
On top of scheduling, Carly acts like a full executive assistant, sending me my schedule the night before so I can prepare for each call. It reminds me of the old x.ai assistant, but Carly is noticeably smarter, faster, and better suited to my healthcare recruitment business."
