Persistent task lists
Keep tasks available across conversations and sessions. A new chat picks up where the last one left off.
Task management built for AI agents
Your AI agents have things to do.
Give them a to‑do list.
A simple, persistent task manager your AI agents can use. Create tasks, organize work, and track progress across conversations.
- Persistent tasks
- Works across sessions
- Built for AI agents
Interactive demo: an AI agent managing a task list
Illustrative demo with sample data. Not connected to any account.
Agent conversation
Connected to TodosCreate a list of everything we need to do before Friday's customer launch.
I've created five tasks in a new Friday launch list and organized them by priority.
Created 5 tasks · Friday launch
Friday launch
The problem
Agents can plan the work. But where does the plan go?
AI agents can draft plans, write checklists, and work through complex jobs. But a task that only exists inside a conversation is hard to find again once that conversation ends.
Problem 1: Tasks get buried in conversations
A checklist written in chat scrolls out of view. Finding it again means searching old threads.
Problem 2: Work is hard to track between sessions
A new session starts with a blank slate. The agent can't see what it promised to do yesterday.
Problem 3: People can't see what agents are doing
Without a shared list, the only way to check progress is to ask the agent, and hope it remembers.
Todos for Agents is the system of record
Tasks live in a hosted list, not in a chat. Any session can read them, update them, and pick up the next one. People can check the same list at any time.
Features
Everything your agents need to stay on task.
Priorities and deadlines
Organize what matters and when it needs to happen, so your agent works on the right thing first.
Task updates
Create, modify, complete, and retrieve tasks. Every change is saved the moment the agent makes it.
Human visibility
Give people a clear view of outstanding and completed work, without reading through chat logs.
Organized work
Group related tasks into lists, one per project, client, or routine, so nothing gets mixed together.
Agent-friendly access
Connect compatible AI agents directly to the service. No custom code or database to manage.
How it works
Connect your agent. Start getting things done.
- 1
Create your workspace
Sign up and get a persistent place for your tasks.
- 2
Connect your AI agent
Connect a compatible agent using the service's integration instructions.
- 3
Put your agents to work
Ask your agent to create tasks, update progress, and retrieve outstanding work.
For developers: how the connection worksModel Context Protocol (MCP), endpoint, and an example client config
Agents reach Todos for Agents through the Model Context Protocol, an open standard for connecting AI assistants to tools. The service runs as a remote MCP server. Your agent discovers its task operations (create, update, complete, list, and so on) when it connects, so there's no SDK to install.
Example only. Your workspace shows your real endpoint and sign-in method. The value below is a placeholder, not a live address.
{
"mcpServers": {
"todos": {
"type": "http",
"url": "https://<your-endpoint-from-the-dashboard>/mcp"
}
}
}Configuration formats differ between clients. The connection guide covers the steps and will list tested clients as they're confirmed.
Want the detail? Read the full walkthrough of how Todos for Agents works.
Use cases
A little organization goes a long way.
Four everyday ways teams put a persistent to-do list in front of their agents.
Sales follow-ups
Track prospects, next steps, and customer commitments.
“Add a follow-up for every call I had this week, due within two days.”
On the list
- Send pricing to Harbor Dental
- Confirm Thursday demo with Ortega Logistics
- Check in on the Bright Path renewal
Business operations
Manage recurring operational tasks and outstanding work.
“Start this month's close checklist and tell me what's still open from last month.”
On the list
- Reconcile vendor invoices
- Renew expiring software licenses
- Send monthly metrics to leadership
Project coordination
Maintain checklists, milestones, and assignments.
“Break the website refresh into tasks and assign the copy work to Priya.”
On the list
- Draft new homepage copy · Priya
- Collect product screenshots
- Review staging site before launch
Personal productivity
Let an AI assistant help manage day-to-day priorities.
“What's the most important thing on my list today?”
On the list
- Renew passport · due Friday
- Book dentist appointment
- Reply to the landlord about the lease
Built for agents, useful for humans
Your agents manage the work. You stay in control.
Agents and people read and write the same task records. Nothing is hidden inside a conversation.
- One set of records
- There's no separate copy for the agent and another for people. Everyone works on the same tasks.
- Every change shows up
- When an agent completes a task, the web view shows it. When you change a priority, the agent sees it next time it checks.
- You decide what matters
- Reorder priorities, reassign work, or add tasks yourself. The agent takes its next task from the list you shaped.
Friday launch
6 tasks · 3 done
Open
2
In progress
1
Done
3
Activity
Agent created 5 tasks in Friday launch
Dana changed the due date on Verify billing configuration
Agent completed Review landing page
Dana added Brief support team on launch FAQs
What is Todos for Agents?
Todos for Agents is a hosted to-do list built for AI agents. Your agent can create tasks, set priorities and due dates, mark work complete, and look up what's still outstanding. The tasks are stored in the service, not in a conversation, so they're still there the next time the agent, or a person, checks.
How is this different from a regular to-do list?
A regular to-do list is designed for people clicking buttons. Todos for Agents is designed for an AI agent to read and update directly, with clear operations for creating, updating, completing, and retrieving tasks. People still get a simple web view of the same tasks, so both sides work from one list.
Can I use Todos for Agents with my existing AI assistant?
Agents connect over the Model Context Protocol (MCP), an open standard for connecting AI assistants to tools and data. If your assistant or agent framework can connect to a remote MCP server, it can connect to Todos for Agents. We'll publish a list of tested assistants and setup steps in the documentation.
Do I need NimbleBrain to use it?
No. Todos for Agents is a standalone product. You don't need a NimbleBrain account or any other NimbleBrain product to use it. NimbleBrain builds and runs the service.
Can people manage tasks too?
Yes. People can view, add, edit, and complete tasks through the web interface. Agents and people work with the same task records, so a change made by one is visible to the other.
Do tasks persist between conversations?
Yes. That's the point of the product. Tasks are saved in the hosted service when they're created or changed, so a new conversation or a different agent session can pick up the same list.
How do I connect an agent?
Create a workspace, then add Todos for Agents to your agent as an MCP server using the connection details from your workspace. The documentation walks through each step.
Is Todos for Agents free?
Pricing hasn't been published yet. Plan details, including whether there is a free tier, will be published on this site.
Give your agents a place to get things done.
Stop losing track of work between conversations. Give your AI agents a persistent to-do list.