Connect Monton to Claude, Cursor, and any MCP-compatible AI assistant. Manage projects, tasks, and time entries from chat.
Replaces CSV exports
Free up to 5 people · No cardBring Monton into your AI workflow
Monton exposes a native MCP server so any compatible LLM can read and act on your projects, tasks, and time — through plain conversation.
Ask anything about your work
Query projects, tasks, hours logged, and team capacity in natural language. Get summaries, status updates, and answers without leaving your AI chat.
Manage projects from chat
Create projects, update tasks, assign people, and move work forward by simply describing what you want done.
Log time conversationally
Tell your assistant what you worked on and let it record time entries on the right project — no forms, no clicks.
Connect your AI app in a minute
Add Monton's server URL to your AI app, sign in and click Allow access. There's no API key to create, paste or rotate.
Sign in instead of pasting a key
Monton connects through OAuth, the same kind of sign-in as “Continue with Google”. Your AI app finds Monton's sign-in page on its own, so all it needs is the server URL.
- 1Add the server URL to your AI app
- 2A browser window opens on Monton
- 3Sign in to your Monton account
- 4Review the permissions and click Allow access
https://the.monton.app/api/mcpStreamable HTTP transport. SSE is not supported.
For claude.ai and Claude Desktop.
- 1Open Settings → Connectors → Add custom connector.
- 2Name it “Monton” and paste the server URL below.
- 3Click Connect, sign in to Monton in the browser and click Allow access.
https://the.monton.app/api/mcpNo API key or headers needed. You'll sign in to Monton in your browser.
What you'll see when you connect
Before any AI app can read or change your data, Monton asks you to approve it.
- Your organization
- The connection uses the organization you're signed in to. The screen shows your email and organization so you can check them before you allow access.
- You choose the permissions
- Every permission is offered and ticked. Untick anything you don't want your assistant to read or change.
- A warning for apps that registered themselves
- These apps show “Only continue if you started this connection”. If you didn't just add Monton to an AI app, click Cancel.
Claude wants to access your Monton account
Signed in as ana@studionorte.com · Studio Norte
Only continue if you started this connection
- TasksRead · Write
- ProjectsRead · Write
- PhasesRead · Write
- Time entriesRead · Write
- Team membersRead
- +4
Permissions
Everything stays within your own organization and within what your Monton account can already do.
| Permission | Access | What your assistant can do |
|---|---|---|
| Tasks | Read · Write | List, read, create and edit tasks, including tags and archiving |
| Projects | Read · Write | List projects and create new ones |
| Phases | Read · Write | Read project phases and create new ones |
| Time entries | Read · Write | Log time and list time entries |
| Team members | Read | List the people in your team |
| Project files | Read | List a project's files |
| Project links | Read | List a project's links |
| Tags | Read · Write | Read and manage tags |
| Project profitability | Read | See how profitable each project is |
Staying connected
- How long does a connection last?
- Access tokens last one hour and your AI app refreshes them automatically. The refresh window is 30 days and starts again every time you use it, so you stay connected as long as you use the app at least once every 30 days.
- How do I disconnect an app?
- In Monton, go to User settings → Connected apps and click Disconnect. Access ends immediately.
Troubleshooting
- “The app's return address does not match”
- Remove the Monton connector from your AI app, add it again and reconnect.
- Connected to the wrong organization
- In Monton, switch to the right organization, disconnect the app under User settings → Connected apps, then reconnect.
- My client doesn't support OAuth
- Create a personal access token in User settings and send it as a Bearer token in the Authorization header. This is the older method, so use OAuth whenever your client supports it.
What is MCP?
The Model Context Protocol (MCP) is an open standard that lets LLMs talk to external systems through a single, secure interface. Connect Monton once and your assistant can plan, query, and update your agency without custom integrations.
- ✓Works with Claude, Claude Code, Cursor, VS Code and any MCP client that supports OAuth
- ✓Secure OAuth sign-in: no API keys to create, paste or rotate
- ✓Read and write projects, phases, tasks and time entries, and see project profitability
- ✓Disconnect any app at any time from User settings → Connected apps
{
"mcpServers": {
"monton": {
"url": "https://the.monton.app/api/mcp"
}
}
}Get connected in 3 steps
- 1
Add the server URL
Paste https://the.monton.app/api/mcp into Claude, Cursor, VS Code or any other MCP client.
- 2
Sign in and allow
A browser opens. Sign in to Monton, review the permissions and click Allow access.
- 3
Start chatting
Ask your assistant to list projects, log time, or update tasks.
Features that work with this one.
What if the follow-up did itself?
Agents that watch budgets, chase invoices and draft client reports, with your approval.
What about the hours from all those meetings?
Turn your calendar events into logged hours in one click.
Log hours without leaving your editor?
Browse projects and tasks and run the timer from VS Code or Cursor.
One flat price. AI agents included. The whole team.
No per-seat fees. Every core feature on every plan, and AI agents on every paid one. When your team grows, your bill doesn't explode.
View plansPrice per plan, not per user · billed annually