Roma has launched Roma 2.0, a redesigned productivity platform designed to help people coordinate multiple artificial intelligence agents through a shared task list. Announced by founder Jan Senderek on November 11, 2026, the updated product aims to solve a growing problem for AI users: different assistants can perform useful tasks, but they often operate independently without sharing a common workspace or understanding what other agents are doing.
The new version positions Roma as a central command center for AI agents, allowing users to delegate tasks and track their progress in one place. It supports connections with tools including ChatGPT, Claude, Grok Bot, Muse and other AI assistants. Rather than replacing these models, Roma provides a shared system where people and their agents can organize work, hand off assignments and review completed tasks.
What Is Roma 2.0?
Roma 2.0 is a productivity application built around the idea of a to-do list for AI agents.
Traditional task-management applications help users record activities, set deadlines and organize projects. However, they generally depend on people to complete the work themselves.
Roma is attempting to change that approach by allowing connected AI agents to interact with the same task workspace.
Users can describe what they need in natural language. Roma turns those instructions into tasks and makes them available to connected agents that can help complete the work.
For example, a user could ask an AI assistant to research a business opportunity, create a task in Roma and allow another connected agent to continue working on it.
The platform is designed to preserve shared context, so users do not have to manually transfer every instruction and update between different AI tools.
Founder Jan Senderek described the product as the first to-do list built specifically for AI agents.
Multiple AI Agents Can Work From One Shared Task List
One of Roma 2.0’s central features is its ability to connect different AI assistants to a common workspace.
The company says users can connect agents such as:
- ChatGPT: OpenAI’s AI assistant.
- Claude: Anthropic’s AI assistant.
- Grok Bot: An AI agent from the Grok ecosystem.
- Muse: An AI agent that can participate in delegated workflows.
- Codex: OpenAI’s coding-focused agent.
- Instinct: Another supported agent integration.
Roma also supports connections with additional tools, including Claude Code, Cursor, Gemini CLI and other applications that support its integration methods.
This approach could be useful for people who prefer different AI assistants for different types of work.
A developer might use Codex for programming, Claude for research and ChatGPT for planning. Instead of maintaining separate task lists, Roma provides a shared place to organize assignments and monitor activity.
The platform’s goal is to make these agents work together without requiring users to manage every handoff manually.
Roma Automatically Turns Instructions Into Tasks
Roma 2.0 is designed to make task creation easier through natural-language input.
Users can describe an activity in ordinary language rather than filling out multiple fields or creating a detailed checklist manually.
The application can convert these instructions into structured tasks and assign them to the appropriate agent.
The system is intended to make the process of delegating work more straightforward.
For example, a user could describe a research assignment, and Roma could create a task that a connected agent can pick up. The user can then return to the application to review the progress and output.
This approach moves task management beyond simply recording what needs to be done. It introduces an execution layer where AI agents can participate in completing the work.
However, the actual results still depend on the capabilities, permissions and integrations of the connected agents.
Every Task Has a Record of Its Progress
Another important feature of Roma is its emphasis on keeping work visible.
The application provides a central location where users can see tasks and follow the activity associated with them.
Roma’s product description highlights a record of the steps taken by agents, creating a shared history of what happened during a task.
The company has also introduced a connection section on each agent’s page. This helps users understand whether an agent can communicate with Roma and whether Roma can reach that agent.
A testing option allows users to check the connection.
This visibility is important because AI agents can sometimes perform actions without users having a clear picture of what happened.
A shared task history could make it easier to review work, identify incomplete assignments and understand which agent handled a particular activity.
Roma’s changelog also highlights a logging feature that records changes to lists and individual items, including information about who made changes and when.
Roma Connects With Everyday Productivity Applications
Roma’s broader product ecosystem extends beyond AI chatbots.
The platform can connect with productivity services such as Gmail, Notion and Google Drive, allowing agents to use information from existing workflows.
These integrations can support tasks such as drafting email replies, finding relevant documents and organizing information.
Roma also provides an MCP server, which allows compatible AI applications to interact with its workspace.
MCP, or Model Context Protocol, is a standard that enables AI applications to connect with external tools and data sources.
Roma’s developer documentation lists support for ChatGPT, Claude, Codex, Muse, Cursor, Gemini CLI and other MCP-compatible applications.
This means users can potentially manage Roma tasks directly through the AI assistant they already use, rather than always opening the productivity application.
Who Is Behind Roma?
Roma was founded by Jan Senderek, who previously founded Loom, a company acquired by Dropbox in 2014.
Senderek announced Roma 2.0 on November 11, 2026, describing it as a shared task-management system for the growing ecosystem of AI agents.
The startup is backed by Y Combinator and operates under the company name Milo Mode Inc.
Roma is part of a broader shift in productivity software, where developers are attempting to build applications that do more than store information or organize schedules.
Instead, these products are increasingly designed to let AI systems execute tasks, interact with external services and coordinate work across multiple applications.
What Makes Roma 2.0 Different From Traditional To-Do Apps?
The biggest difference is that Roma is designed around delegation to AI agents rather than only personal task organization.
| Feature | Traditional to-do apps | Roma 2.0 |
|---|---|---|
| Task creation | Primarily manual | Natural-language input |
| Task execution | Usually handled by users | Connected AI agents can perform work |
| Multiple AI agents | Generally not central to the product | Shared agent workspace |
| Task tracking | Status and deadlines | Agent activity and task history |
| Integrations | Productivity services | Productivity services and AI agents |
| Collaboration | People and teams | People, teams and connected agents |
This positioning could become increasingly relevant as individuals and businesses begin using multiple AI assistants simultaneously.
However, Roma is not an AI model itself. It does not replace Claude, ChatGPT or other underlying systems. Its role is to coordinate their work through a shared productivity environment.
The Bigger Picture
Roma 2.0 arrives as the AI industry moves from conversational assistants toward autonomous agents capable of completing multi-step tasks.
Until recently, most users interacted with AI tools individually. They opened a chatbot, entered instructions, received an answer and moved to another application to continue working.
The emergence of AI agents is changing that workflow. People increasingly want software that can research, organize information, draft documents and carry out actions across different services.
Roma is betting that managing these agents will become a productivity challenge of its own. A shared task list could provide a practical way to coordinate different assistants without forcing users to depend on a single AI provider.
The opportunity is significant, but the platform will need to demonstrate reliable integrations, clear permission controls and consistent task execution before it can become an essential tool for everyday work.
Looking Ahead
Roma’s next challenge will be expanding its agent ecosystem and making connections between different AI assistants reliable enough for regular use. Its success will depend on whether users find it easier to delegate work through a shared task manager than to coordinate their agents independently. The quality of completed tasks, integration reliability and transparency around automated actions will be important factors.
More broadly, Roma 2.0 reflects a potential change in how people use productivity software. As AI agents become more capable, the value of an application may increasingly depend on how effectively it coordinates work rather than simply helping users organize it. Roma is positioning itself around that possibility by turning the traditional to-do list into a shared workspace for people and AI agents.
SEO Title: Roma 2.0 Launches as a Shared To-Do List for ChatGPT, Claude and AI Agents
Meta Description: Roma 2.0 connects ChatGPT, Claude, Grok and other AI agents through a shared task list, allowing users to delegate work and track progress.
URL Slug: roma-2-launches-shared-to-do-list-ai-agents
Focus Keyphrase: Roma 2.0 launch
Secondary Keywords: Roma AI app, Jan Senderek, AI agent productivity, shared AI task list, ChatGPT Claude integration, AI agent management, Roma MCP server, AI automation tools
Get the day’s top stories in your inbox
One concise email. No spam, unsubscribe anytime.
