Allows agents to create and manage tasks in Todoist, including setting descriptions, due dates, priorities, labels, and projects, as well as listing and filtering active or completed tasks.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@Todoist MCP Server Add a task 'Review project proposal' due tomorrow with high priority"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Unofficial MCP server for Todoist that allows agents to create, and list tasks in your Todoist account.
Create tasks with descriptions, due dates, priorities, and labels and projects.
List tasks List completed or uncompleted tasks with filtering by project or Todoist filters.
Works with Claude Desktop, Cursor, and other MCP clients
Related MCP server: Todoist MCP Server
pip install todoist-mcp-serverScroll down to "API token"
Copy your API token (keep it secure!)
Here are some examples of how to use the Todoist MCP server with different clients:
Add this to your claude_desktop_config.json file: ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%/Claude/claude_desktop_config.json (Windows)
{
"mcpServers": {
"todoist": {
"command": "todoist-mcp-server",
"env": {
"TODOIST_API_TOKEN": "your-api-token-here"
}
}
}
}Add this to your Cursor settings:
Open Cursor Settings (Cmd/Ctrl + ,)
Search for "MCP"
Add the MCP server configuration:
{
"mcpServers": {
"todoist": {
"command": "todoist-mcp-server",
"env": {
"TODOIST_API_TOKEN": "your-api-token-here"
}
}
}
}For any MCP-compatible client, use:
Command: todoist-mcp-server
Environment Variable: TODOIST_API_TOKEN=your-token
Restart Claude Desktop, Cursor, or your MCP client to load the server.
Once configured, you can interact with Todoist using natural language:
"Create a task to buy groceries"
"Add a task 'Call dentist' due tomorrow with high priority"
"Create a task to finish the report with description 'Include Q4 metrics' due next Friday"
"Show me my tasks for today"
"List all my high priority tasks"
"What tasks do I have in my Work project?"
create_taskCreate a new task in Todoist.
Parameters:
content (required): Task title/content
description (optional): Task description
project_name (optional): Project name to add task to
due_string (optional): Due date in natural language ("tomorrow", "next monday")
priority (optional): Priority level 1-4 (1=low, 2=medium, 3=high, 4=urgent)
labels (optional): List of label names
list_active_tasksList active tasks from Todoist.
Parameters:
project_name (optional): Filter by project name
filter_string (optional): Todoist filter ("today", "overdue", "p1")
limit (optional): Maximum number of tasks (default: 50)
Make sure you've installed the package: pip install todoist-mcp-server
Verify your API token is correct
Check that the config file is in the right location
Restart your MCP client completely
The most common cause is that your MCP client can't find the todoist-mcp-server command in its PATH.
Solution: Use the full path to the command
Find where the command is installed:
which todoist-mcp-serverUse the full path in your MCP config:
{
"mcpServers": {
"todoist": {
"command": "/full/path/to/todoist-mcp-server",
"env": {
"TODOIST_API_TOKEN": "your-api-token-here"
}
}
}
}Other fixes:
Ensure the package is installed in the same Python environment your MCP client uses
Try reinstalling: pip uninstall todoist-mcp-server && pip install todoist-mcp-server
Get a fresh token from Todoist Integrations
Make sure there are no extra spaces in your config file
Verify the token has the necessary permissions
To contribute or run from source:
git clone https://github.com/mehularora8/todoist-mcp
cd todoist-mcp
pip install -e .Your API token is stored locally and only used to communicate with Todoist's API
No data is sent to third parties
The MCP server runs locally on your machine
MIT License - see LICENSE file for details.
🐛 Issues: Report bugs or request features
📖 Todoist API: Official Documentation
🔧 MCP Protocol: Model Context Protocol
Note: This is an unofficial integration and is not affiliated with Todoist or Doist Inc.
Create a new task in Todoist.
Priorities don't map directly to Todoist priorities, but are mapped as follows:
1 = low
2 = medium
3 = high
4 = urgent
This is done to map better to how people think about priorities.
Args:
content: The task content/title (required)
description: Task description (optional)
project_name: Project name to add task to
due_string: Due date in natural language like "tomorrow", "next monday" (optional)
priority: Priority level 1-4 (1=low, 2=medium, 3=high, 4=urgent)
labels: List of label names to add to the task (optional)
Returns:
Dict containing the created task details or error message
| Name | Required | Description | Default |
|---|---|---|---|
| content | Yes | ||
| description | No | ||
| project_name | No | ||
| due_string | No | ||
| priority | No | ||
| labels | No |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries the full burden of behavioral disclosure. It clearly indicates this is a write operation ('Create'), explains the priority mapping system in detail, and mentions the return format ('Dict containing the created task details or error message'). However, it doesn't address authentication requirements, rate limits, or what happens with invalid inputs.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is well-structured with clear sections (purpose, priority mapping, args, returns) and every sentence adds value. The priority mapping explanation is necessary but slightly lengthy. Overall efficient with minimal waste, though could be slightly more concise in the priority explanation.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a creation tool with no annotations and no output schema, the description provides substantial context: clear purpose, detailed parameter semantics, return format indication, and priority system explanation. The main gaps are lack of error handling details and authentication context, but overall it's quite complete for the tool's complexity.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
With 0% schema description coverage, the description fully compensates by providing detailed semantic explanations for all 6 parameters. Each parameter gets clear explanations of purpose, optionality, format expectations (e.g., 'natural language like "tomorrow"'), and the priority mapping table adds crucial context not inferable from the schema alone.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the specific action ('Create a new task') and resource ('in Todoist'), distinguishing it from sibling tools like list_active_tasks and list_completed_tasks which are read operations rather than creation operations. The verb+resource combination is precise and unambiguous.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies usage context through the parameter explanations (e.g., 'optional' flags), but doesn't explicitly state when to use this tool versus alternatives. No guidance is provided about prerequisites, error conditions, or comparison with sibling tools beyond their different functions.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
List tasks from Todoist
Args:
project_name: Filter tasks by project name (alternative to project_id)
filter_string: Todoist filter string like "today", "overdue", "p1" (optional)
limit: Maximum number of tasks to return (default 50)
Returns:
Dict containing list of tasks or error message
| Name | Required | Description | Default |
|---|---|---|---|
| project_id | No | ||
| project_name | No | ||
| filter_string | No | ||
| limit | No |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries the full burden of behavioral disclosure. It mentions the tool lists tasks and returns a dict with tasks or error, but lacks details on permissions, rate limits, pagination, or what constitutes an 'active' task. For a read operation with zero annotation coverage, this leaves significant behavioral gaps.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is well-structured with clear sections (Args, Returns) and uses bullet-like formatting efficiently. Each sentence adds value, though the omission of 'project_id' in Args slightly reduces clarity. Overall, it's appropriately sized and front-loaded with the core purpose.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given 4 parameters with 0% schema coverage and no output schema, the description compensates well for parameters but lacks output details (only mentions 'Dict containing list of tasks or error message'). For a tool with no annotations and moderate complexity, it's adequate but incomplete, especially regarding behavioral context and output structure.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The description adds substantial meaning beyond the input schema, which has 0% description coverage. It explains that 'project_name' filters by project name as an alternative to 'project_id', provides examples for 'filter_string' ('today', 'overdue', 'p1'), and specifies the default for 'limit' (50). However, it omits 'project_id' from the Args section, creating a minor inconsistency.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the verb ('List') and resource ('tasks from Todoist'), making the purpose immediately understandable. It distinguishes from 'list_completed_tasks' by specifying 'active' in the tool name, though the description doesn't explicitly contrast with siblings beyond the name.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies usage through parameter explanations (e.g., 'alternative to project_id'), suggesting when to use certain filters, but lacks explicit guidance on when to choose this tool over 'list_completed_tasks' or 'create_task'. No clear when-not-to-use or alternative tool recommendations are provided.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
List completed tasks from Todoist within a timespan. Default is last 24 hours.
Prefer not to provide `since` and `until` if you want to pull tasks from the last 24 hours.
Args:
project_name: Filter tasks by project name (optional)
since: Start date in ISO format (YYYY-MM-DD) in the user's timezone (optional)
until: End date in ISO format (YYYY-MM-DD) in the user's timezone (optional)
limit: Maximum number of tasks to return (default 30, max 200)
Returns:
Dict containing list of completed tasks or error message
| Name | Required | Description | Default |
|---|---|---|---|
| project_name | No | ||
| since | No | ||
| until | No | ||
| limit | No |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries full burden. It discloses the default time range (last 24 hours), default limit (30), and maximum limit (200), which are useful behavioral traits. However, it doesn't mention authentication requirements, rate limits, pagination behavior, or what happens when no tasks match criteria.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is well-structured with purpose statement, usage guidance, parameter explanations, and return value description. Every sentence adds value, though the 'Returns' section could be slightly more specific about the dict structure. It's appropriately sized for a 4-parameter tool.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a read operation with 4 parameters and no output schema, the description provides good coverage of purpose, parameters, and basic behavior. It lacks details about authentication, error handling specifics, and exact return structure, but given the tool's relative simplicity and the parameter documentation provided, it's mostly complete.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
With 0% schema description coverage, the description fully compensates by explaining all 4 parameters. It clarifies that project_name is optional and for filtering, specifies ISO format and timezone context for since/until, explains the default and maximum values for limit, and provides usage guidance about when to omit since/until parameters.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the verb ('List'), resource ('completed tasks from Todoist'), and scope ('within a timespan'). It explicitly distinguishes from the sibling 'list_active_tasks' by specifying 'completed tasks' and provides temporal context that differentiates it from 'create_task'.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides clear context about when to use default behavior ('Prefer not to provide `since` and `until` if you want to pull tasks from the last 24 hours'), but doesn't explicitly mention when to use this tool versus the sibling 'list_active_tasks' or 'create_task'. It offers guidance on parameter usage but lacks explicit sibling comparison.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
v0.1.3create_tasklist_active_taskslist_completed_tasksScored across 3 tools
Each tool has a clearly distinct purpose: create_task for creating new tasks, list_active_tasks for retrieving current tasks, and list_completed_tasks for viewing past tasks. There is no overlap in functionality, making it easy for an agent to select the right tool.
All tool names follow a consistent verb_noun pattern (create_task, list_active_tasks, list_completed_tasks) with clear and descriptive naming. There are no deviations in style or convention.
With only 3 tools, the set feels too thin for a comprehensive Todoist integration. Key operations like updating tasks, deleting tasks, managing projects, or marking tasks as complete are missing, limiting the server's utility for full task management workflows.
The tool set is severely incomplete for a Todoist server. While it covers creation and listing (active and completed), it lacks essential CRUD operations such as update_task, delete_task, and complete_task, as well as project and label management tools, creating significant gaps that will hinder agent performance.
Official Todoist MCP server for AI assistants to manage tasks, projects, and workflows.
Nifty's MCP server — exposes tasks, projects, messages, and files as tools for AI agents.
Manage Superlist tasks and lists in plain language from any MCP-compatible AI agent.
