Skip to content
Todos for Agentsby NimbleBrain
Get Started

How it works

The list lives outside the conversation.

Your agent talks to Todos for Agents the way it talks to any other tool. The tasks stay in the service, so every new session starts from the current list, not a blank page.

Three steps to set up

  1. 1

    Create your workspace

    Sign up and get a persistent place for your tasks.

  2. 2

    Connect your AI agent

    Connect a compatible agent using the service's integration instructions.

  3. 3

    Put your agents to work

    Ask your agent to create tasks, update progress, and retrieve outstanding work.

Step-by-step instructions are in the connection guide.

How tasks survive between sessions

Most agents forget everything when a conversation ends. With Todos for Agents, the agent writes tasks to the service as it works, then reads them back at the start of the next session.

That also works across agents. Two sessions, or two different agents connected to the same workspace, see the same lists.

Monday session

creates 5 tasks

Todos for Agents

the saved list

Tuesday session

reads 4 still open

You, in the web view

see and edit the same list any time

The life of a task

Status is explicit. An agent doesn't guess whether something is finished; it marks it. Tasks can also be cancelled when they're no longer needed.

  1. 01

    Open

    The agent (or a person) adds the task with a priority, a due date, and an assignee if there is one.

  2. 02

    In progress

    The agent marks what it's working on, so a second session doesn't start the same thing.

  3. 03

    Blocked

    If the agent is waiting on something, it records the reason. People can see what needs their input.

  4. 04

    Done

    The agent marks it complete. Anyone checking the list sees the change right away.

Who does what

Responsibilities of the agent, people, and the service
RoleResponsible for
Your agentDoing the work, then creating, updating, and completing tasks as it goes.
Your teamSetting priorities, reviewing progress, and stepping in when something is blocked.
Todos for AgentsStoring the tasks, keeping them in sync for everyone, and serving them to any connected agent.

The technical detail

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.

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.

Get StartedRead the docs