MCP setup

Use Done in Claude

Connect Done to Claude as a custom connector: paste https://trydone.app/mcp, sign in with OAuth, and Claude can read, plan and work your tasks — with your approval.

Who this is for

For anyone using Claude on the web, in Claude Desktop or in Cowork, on any plan — the Free plan allows one custom connector. Claude reaches Done from Anthropic's cloud, so nothing runs on your computer, and the connection uses your own Done sign-in: there's no key to copy.

Set it up

Done's MCP server URL

https://trydone.app/mcp

Add Done as a custom connector Sign in with OAuth

In Claude, open Customize → Connectors and choose Add custom connector.
Name it Done and enter https://trydone.app/mcp as the server URL.
If Claude asks how people sign in, choose Sign in now. If it asks for an OAuth client, choose Register automatically, and leave any client ID and secret empty — Claude registers itself with Done.
Click Add, then Connect. Claude opens Done: sign in, check the agent's name, and click Allow.
In a chat, open the + menu → Connectors and make sure Done is switched on.

On a Team or Enterprise plan, an Owner adds the connector once under Organization settings → Connectors. Each member then clicks Connect next to it under Customize → Connectors and signs in with their own Done account.

Or send a token as a request header MCP token

In Done, open Settings → Agent access and generate an MCP token. It's shown once, so copy it straight away.
In Claude's Add custom connector dialog, choose No sign-in, open Request headers, pick authorization, and enter Bearer followed by your token — Claude sends the value exactly as typed.

Request headers are in beta in Claude and not every organization has them yet. If you don't see them, use the sign-in above.

Need a narrower key? Under Settings → Agent access → Extra MCP tokens you can create one that's read-only, or limited to one project or one workspace — the agent then only ever sees that much.

What to say to it

Once Done is connected, talk to the assistant normally. Each of these maps to a real Done tool:

“What's on my plate today?”

get_todays_plan

“Add a task: send the invoice to Sam friday p1”

quick_capture

“Pick up your next Done task and propose a plan before you start.”

next_task → propose_plan

“If you hit a decision you can't make, ask me and keep going with the rest.”

ask_decision

“When it's finished, send it to me for approval.”

propose_completion

Every tool, prompt and resource Done exposes is listed in the MCP reference.

The approval gate

By default an assistant can't close your tasks. When it thinks a task is finished it calls propose_completion, which parks the task as Pending approval until you approve it or send it back from Needs you; substantial work starts with propose_plan, which waits for your go-ahead. A question goes through ask_decision: that task waits for your answer while the agent moves on to the next one. Done enforces this on its own server, so it holds whatever the assistant tries — an agent can only complete a task directly when you've set that task's Autonomy to let it.

Troubleshooting

Connecting fails right after you click Connect: If you chose Use Claude's published identity as the OAuth client, remove the connector and add it again with Register automatically. The published identity needs Client ID Metadata Documents, which Done doesn't support yet.
Claude answers without using Done: Check the connector is switched on for that conversation (+ → Connectors), or name it: “Use Done to show my open tasks.”
You want to change the URL or switch to a token: Claude can't edit a connector's authentication after it's added. Remove it under Customize → Connectors and add it again.

FAQ

Can Claude work my Done list on a schedule?

Yes. A Claude routine runs in Anthropic's cloud with Done attached as a connector; the MCP docs have the steps and the prompt to give it.

Is this the same as Claude Code?

No. Claude Code, the coding agent, is set up with one terminal command — see Use Done in Claude Code.

Is there a shared API key for Done?

No. Every connection belongs to one Done account: an OAuth sign-in, or an MCP token you generate yourself. There is no shared or global key. Each connected assistant is listed under Settings → Agent access, where you can revoke it in a click.

What can the assistant see?

Only what your account can see: your workspaces and anything shared with you. A read-only, one-project or one-workspace token narrows that further.

Does it cost anything?

The Free plan includes 100 agent (MCP) calls a month; Pro and Teams are unlimited. The assistant's own plan is separate.

Steps checked against the official documentation on October 23, 2026: