get_tasks_toolLABSCO SUMMARY
Your task list where the work is, in the query language you already use.
Because it takes Todoist's own filter syntax rather than a reduced set of options, any view you have already built in Todoist is a question you can ask from the editor.
What it is
A read view of Todoist built for Cursor. Tasks come back through Todoist's filter query language and are formatted with priority and due date rather than as raw fields.
What you get
- Tasks retrieved with any Todoist filter query — today, tomorrow, overdue, a named date, or a date range
- Filtering by priority level 1 to 4, where 1 is the highest
- Complex queries combined, so "due before May 5 and high priority" is a single call
- Each task rendered with its priority, due date and other relevant fields marked with icons instead of run together
Requirements
Python 3.10 or higher and a clone of the repository, with its dependencies installed into a .venv. A Todoist account and its API token, which you take from Todoist settings under Integrations → Developer and hand to the server as an environment variable. It reads tasks; nothing here creates or completes them.
Setup effort
Build from source — git clone https://github.com/mingolladaniele/taskMaster-todoist-mcp.git
USE CASES
01
Seeing what is due today without switching apps
One filtered read while you are still in the editor, formatted with priorities and dates rather than as raw fields.
02
Catching what slipped
The overdue filter on its own — the list nobody wants to open the app for.
03
Picking the next thing to work on
Filter to priority 1 and today, and the answer is a short list instead of a whole project.
04
Planning a week from a date range
A due-before query returns everything landing ahead of the date you name.
1 tool
Listed in project docs. Your client may show a different list.
BEFORE YOU INSTALL
What to know first
Tools you get
1 tool named in the project's documentation. get_tasks_tool only reads; creating, completing and deleting tasks are planned, not shipped.
Requires
- Python 3.10 or later
- A Todoist API token (TODOIST_API_TOKEN)
Where it runs
The documented client runs the server on your machine.
INSTALLATION
Connect your client
You clone this one yourself.
Server setup
Clone the repository:
TERMINAL
git clone https://github.com/mingolladaniele/taskMaster-todoist-mcp.git
cd taskMaster-todoist-mcpCreate a virtual environment:
TERMINAL
python -m venv .venvActivate it on macOS or Linux:
TERMINAL
source .venv/bin/activateOr on Windows:
TERMINAL
.venv\Scripts\activateInstall the dependencies:
TERMINAL
pip install "mcp[cli]==1.5.0" requests python-dotenvGet your Todoist API token:
In Todoist, open Settings → Integrations → Developer and copy your API token; the client configuration passes it as TODOIST_API_TOKEN
MCP Clients
Cursor
CURSOR · JSON
{
"mcpServers": {
"todoist-mcp": {
"command": "C:/Users/<YOUR_USERNAME>/<YOUR_PATH_TO_TODOIST_MCP>/.venv/Scripts/python.exe",
"args": [
"C:/Users/<YOUR_USERNAME>/<YOUR_PATH_TO_TODOIST_MCP>/server.py"
],
"env": {
"TODOIST_API_TOKEN": "<YOUR_TODOIST_API_TOKEN>"
}
}
}
}This is the README's Windows example. C:/Users/<YOUR_USERNAME>/<YOUR_PATH_TO_TODOIST_MCP> is your clone's folder. On macOS or Linux the interpreter is .venv/bin/python inside it. Put your Todoist API token in TODOIST_API_TOKEN.
