Todoist MCP. Manage tasks, projects, and comments right from your chat.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Just plug in your AI agents and start using Vinkius.
Todoist MCP Server connects your Todoist account directly to any AI client. You can view active tasks, manage projects, create new items, and check discussion history—all without ever having to open the native app.
It lets you keep complex project data inside your chat window.
What your AI agents can do
Complete task
Marks an existing task as finished in Todoist.
Create task
Adds a brand new item to your task list, specifying content and optionally a project ID.
List comments
Retrieves all comments associated with a specific task.
Retrieves a list of every active project folder in your Todoist workspace.
Lists all sections contained within a specific, designated project.
Retrieves a list of active tasks, with the option to filter by a specific project ID.
Adds a new task item to a chosen project or your general inbox using plain conversational input.
Changes the status of an existing, pending task to completed immediately.
Retrieves all discussion history and comments associated with a specific task.
Provides a full list of user-defined labels available for categorizing your work.
Ask AI about this MCP
Supported MCP Clients
OAuth 2.0 Compatible Gemini Waiting for input…
Todoist MCP Server: 7 Tools for Task & Project Ops
These seven tools let your AI agent perform every core action in Todoist—from listing projects to completing tasks—all without you leaving the chat.
Make your AI actually useful.
Add this MCP to Claude, Cursor, or Windsurf and your AI stops guessing. It gets real tools to look things up, take action, and handle the stuff you keep doing by hand.
Start using Todoist on Vinkius019d7613complete task
Marks an existing task as finished in Todoist.
019d7613create task
Adds a brand new item to your task list, specifying content and optionally a project ID.
019d7613list comments
Retrieves all comments associated with a specific task.
019d7613list labels
Shows every user-defined label or tag you have set up in your account.
019d7613list projects
Retrieves a list of all currently active project folders.
019d7613list sections
Lists the specific organizational sections inside one given project.
019d7613list tasks
Shows your pending tasks, with an option to filter them by a project ID.
Choose How to Get Started
Build a custom MCP for your own tools, or connect a ready-made integration from our catalog.
Build Your Own
Turn any API into an MCP. Import a spec, define Agent Skills, or deploy with MCPFusion.
- Import from OpenAPI, Swagger, or YAML specs
- Create Agent Skills with progressive disclosure
- Deploy to edge with MCPFusion framework
- Built in DLP, auth, and compliance on every call
- Real time usage dashboard and cost metering
- Publish to catalog or keep private
Make Your AI Do More
Start with Todoist, then connect any of our 4,800+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 4,800+ others, all in one place
- Add new capabilities to your AI anytime you want
- Every connection is secured and compliant automatically
- Track usage and costs across all your servers
- Works with Claude, ChatGPT, Cursor, and more
- New servers added to the catalog every week
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Todoist. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.
VINKIUS INFRASTRUCTURE
Cloud Hosted
Managed infra
V8 Isolated
Sandboxed per request
Zero-Trust Proxy
No stored credentials
DLP Enforced
Policy on every call
GDPR Compliant
EU data residency
Token Compression
~60% cost reduction
Works with Claude, ChatGPT, Cursor, and more
The Model Context Protocol standardizes how applications expose capabilities to LLMs. Instead of operating in isolation, your AI gains direct access to external platforms, live data, and real-world actions through secure, standardized connections.
This server provides 7 capabilities that interface natively with Claude, ChatGPT, Cursor, and any MCP client. No middleware. No custom integration required.
Context switching shouldn't mean opening five different tabs.
Today, getting an overview means logging into Todoist, finding the main project dashboard. Then you click on 'Sections,' which takes you to another page, and if you want to see pending tasks, you have to scroll through a list that might be buried under old comments or completed items. It's clicking, opening, navigating—a painful sequence just to get status.
With this MCP server, the process changes entirely. You simply tell your agent: 'Give me all active projects and their pending tasks.' The agent runs `list_projects` and then feeds those IDs into `list_tasks`. You get a structured summary right in your chat—no tabs to click, no pages to scroll through.
Todoist MCP Server: Get project data instantly.
The manual steps that disappear are the initial login, navigating between Project and Section views, and then manually copying task IDs to check their status. These little friction points add up fast when you're trying to keep track of a complex client scope.
Now, checking project status is one command. The agent runs `list_tasks` with filtering logic built in, returning the exact data set you need—and nothing else. It’s just faster and less error-prone.
What you can do with this MCP connector
You connect Todoist straight to your agent, and you keep all your damn project work right inside this chat window. You don't gotta open the whole Todoist app; we handle it here. It lets you manage projects, check off tasks, and track down discussion history using plain commands.
Project Organization
You can start by seeing every single active folder in your entire workspace with list_projects. That gives you a rundown of all the major project areas you're working on. Once you pick a specific project, you can drill down and view exactly what sections are inside it using list_sections. This shows you how everything is organized within that one project.
Managing Your Tasks
The core function here is task control. You can pull up all your pending items by running list_tasks, and you'll even get the option to filter those tasks down to a single, specific project ID if you know what you're looking for. If you spot something that needs to be finished, just run complete_task on it; the agent marks that task as done immediately in Todoist.
Need to add something? You can create brand new items using create_task. You tell the agent what the task is, and you decide if it belongs in a specific project or if it's just floating in your general inbox. It handles the placement for ya.
Context and Details
If you need context on an old assignment, you can retrieve all discussion history associated with any given item using list_comments. This pulls up every comment made on that task so you don't lose track of what was decided. You can also get a full inventory of your work categories by running list_labels, which shows you every user-defined tag or label you've set up in Todoist.
019d7613-7199-70b5-9c11-a037681ba3cc 
How Todoist MCP Works
- 1 Subscribe to the Todoist MCP Server on Vinkius.
- 2 Go to your Todoist integrations page and generate a personal API Token.
- 3 Input that token into your agent's setup. Now you can run task-management queries naturally.
The bottom line is, once the token is set up, your AI client handles all the complex calls to keep your tasks organized in chat.
Who Is Todoist MCP For?
Anyone whose workflow involves tracking details across multiple projects—especially if they hate context switching. This is for the project manager who needs an immediate status report on 20 different initiatives, or the software developer who wants to check off a coding chore without leaving their IDE. It works when your job requires knowing 'what's next' instantly.
Uses list_projects and list_tasks to gather status reports across departments, keeping track of due dates without opening the web app.
Runs complete_task after finishing a bug fix or feature implementation directly from their terminal session.
Checks the history on key items using list_comments to ensure all stakeholders are aligned before a major release.
What Changes When You Connect
- Instant Status Checks: Need to know what's pending? Use
list_tasksorlist_projectsto pull a full overview of your workload without switching tabs. You get the list immediately in your conversation window. - Zero Context Switching: Don't open Todoist. Just tell your agent, 'What are my tasks for Project X?' The agent runs
list_tasksand gives you the answer instantly. No app required. - Maintain Discussion History: When a task gets complicated, use
list_comments. You can pull up every discussion point on an old item without having to click through pages of chat logs. - Quick Task Closure: Finished working on that chore? Use
complete_taskwith the ID. It instantly marks it done in Todoist—no manual clicking needed. - Structured Project View: You can get a bird's eye view by first running
list_projects, then narrowing down to specific sections usinglist_sections. This lets you map out your entire project structure quickly.
Real-World Use Cases
The Weekly Check-in
A PM needs an immediate status report for a client meeting. Instead of opening the app and clicking through five different projects, they ask their agent to run list_projects first, then loop through them using list_tasks. The agent gathers all pending items into one conversational summary.
Catching Up on a Feature
A developer jumps back into a bug fix they started last week. They need to know what decisions were made. Rather than digging through emails, they ask the agent about the task and use list_comments to see the full discussion history instantly.
Adding an Off-the-Cuff Idea
A quick thought pops up during a meeting. The user doesn't want to open Todoist, so they instruct their agent to run create_task for 'Follow up on Q3 budget' and assign it to the 'Finance' project using plain text.
Cleanup After Deployment
The team just finished a big push. A developer asks the agent to check off all the small, completed coding chores by running complete_task for several IDs, clearing their backlog in one go.
The Tradeoffs
Assuming one command covers everything
Thinking that a single prompt like 'Tell me about the project' will give all necessary data (tasks, sections, comments). It won't.
→
You have to break it down. First, run list_projects to get the scope. Then, use list_sections on the right project, and finally, call list_tasks for the specific section ID. Layering calls works.
Ignoring existing context
Just asking 'What's wrong with task 123.' without giving enough detail to pull up history. The agent needs more data.
→
Always run list_comments when you need the full story on a task ID. It pulls the entire discussion log for context.
Forgetting required IDs
Asking to list tasks without specifying which project they belong to, resulting in an incomplete or unusable list.
→
If you want filtered results, always get the Project ID first using list_projects, then pass that ID into list_tasks for accurate filtering.
When It Fits, When It Doesn't
Use this server if your primary goal is to manage structured tasks and projects within a chat environment. If you need visibility across multiple resource types (Projects, Sections, Tasks), the modularity provided by tools like list_projects, list_sections, and list_tasks makes it ideal for comprehensive overviews.
Don't use this if your needs are simpler—for example, if you only track simple reminders that don't belong to a formal project structure. In that case, a dedicated basic checklist tool might be better. Also, if you need real-time collaboration on the content of a file (like Google Docs), this is for task management, not document editing. If your workflow involves complex cross-resource transactions—meaning 'Update Project A AND Task B simultaneously'—you may need to build a custom wrapper layer over these tools.
Common Questions About Todoist MCP
How do I list all active projects using Todoist MCP Server? +
You run list_projects. This tool fetches every project folder that is currently set up in your account, giving you a clean list of available scopes.
Can I check task comments with Todoist MCP Server? +
Yes. Use the list_comments tool and provide the specific task ID. It retrieves the entire discussion history for that single item, so you don't have to dig through emails.
How do I add a new task in Todoist MCP Server? +
Use create_task. You just need to provide the content of the task and an optional project ID. The agent handles writing it into your todo list for you.
What is the difference between listing tasks and listing projects with Todoist MCP Server? +
list_projects gives you the top-level containers (the folders). list_tasks only shows the actual items inside those containers, usually requiring a project ID to filter correctly.
What steps are needed to authenticate with Todoist MCP Server? +
You must generate a personal API Token from your Todoist account's integration page. This token authorizes your AI client to execute commands like list_tasks and complete_task within your workspace.
Can the Todoist MCP Server filter my active tasks using specific IDs or projects? +
Yes, you can use the list_tasks tool by providing a specific project ID. This filters out irrelevant data instantly, giving you a focused list of only the items relevant to that single project.
Does the Todoist MCP Server have rate limits or handle API errors? +
The server follows standard Todoist API rate limits. If your agent exceeds them, it receives an explicit error message telling you exactly when and why the command failed, preventing data loss.
How does using `list_labels` help me manage task organization? +
Running list_labels shows all your existing user-defined tags. You can then use these labels to categorize and group tasks across different projects, providing a high-level view of your work structure.
Where do I get my Todoist API Token? +
You can locate your personal API Token securely via the web application by clicking your profile avatar, going to Settings, and selecting the Integrations tab. Your token will be visible at the bottom or in a dedicated Developer section. Alternatively, navigate directly to https://todoist.com/prefs/integrations.
Can the agent delete projects or workspaces? +
No. The integration exclusively lists resources (projects, items, labels) and supports only basic item modifications: creating tasks and marking tasks as completed. Destructive actions on structural settings are not supported.
Can the agent organize my tasks using sub-projects or sections? +
Yes. The Todoist integration allows the agent to retrieve project IDs and navigate into their specific sections. When creating a new task, you can explicitly ask the agent to place it within a specific section to keep your work neatly organized.
Multi-server workflows that include Todoist MCP
Use it with your favorite AI tools
Connect this server to Cursor, Claude, VS Code, and more.