OmniFocus MCP Server
An MCP server that enables AI assistants to interact with OmniFocus on macOS via JXA, supporting task, project, folder, tag, perspective, and search operations.
README
OmniFocus MCP Server
A Model Context Protocol (MCP) server that enables AI assistants to interact with OmniFocus on macOS via JXA (JavaScript for Automation).
Features
This MCP server provides access to OmniFocus functionality:
Task Management
- List inbox tasks - View and filter tasks in your inbox
- Create tasks - Add new tasks with full property support (due dates, planned dates, tags, notes, etc.)
- Complete/Drop tasks - Mark tasks as done or dropped
- Get due tasks - Find tasks due within a timeframe
- Get planned tasks - Find tasks planned within a timeframe
- Get flagged tasks - List all flagged items
- Add/remove tags from tasks - Manage task tags
Project Management
- List projects - View projects with status filtering
- Get projects for review - Find projects needing review based on next review date
- Mark project reviewed - Update a project's review status and next review date
- Batch mark reviewed - Efficiently review multiple projects at once
Organization
- List folders - View folder hierarchy
- List tags - View all tags
- List perspectives - View built-in and custom perspectives
- Get perspective tasks - List tasks shown in a specific perspective
Search
- Universal search - Search across tasks, projects, folders, and tags
Requirements
- macOS (OmniFocus is macOS/iOS only, and this server uses JXA)
- OmniFocus 3+ installed
- Node.js 18+
- Automation permissions enabled for your terminal/client app
Installation
-
Clone or download this repository:
cd omnifocus-mcp-server -
Install dependencies:
npm install -
Build the TypeScript:
npm run build -
Configure your MCP client to use the server (see Configuration below)
Configuration
Claude Desktop
Add to your Claude Desktop configuration file (~/Library/Application Support/Claude/claude_desktop_config.json):
{
"mcpServers": {
"omnifocus": {
"command": "node",
"args": ["/path/to/omnifocus-mcp-server/dist/index.js"]
}
}
}
Other MCP Clients
The server uses stdio transport, so configure your client to spawn:
node /path/to/omnifocus-mcp-server/dist/index.js
Permissions
On first use, macOS will prompt you to allow automation access:
- Go to System Preferences → Security & Privacy → Privacy → Automation
- Enable permission for your terminal or Claude Desktop to control OmniFocus
Tool Reference
omnifocus_list_inbox
List tasks in the inbox.
{
"includeCompleted": false,
"limit": 50
}
omnifocus_list_projects
List projects with filtering.
{
"status": "active",
"folderName": "Work",
"limit": 50
}
omnifocus_list_folders
List all folders.
{
"status": "active",
"limit": 50
}
omnifocus_list_tags
List all tags.
{
"status": "active",
"limit": 50
}
omnifocus_list_perspectives
List perspectives (built-in and custom).
{
"limit": 50
}
omnifocus_get_perspective_tasks
Get tasks shown in a specific perspective.
{
"perspectiveName": "Next",
"limit": 50
}
omnifocus_create_task
Create a new task.
{
"name": "Review quarterly report",
"note": "Check all sections",
"projectName": "Work",
"dueDate": "2024-12-31T17:00:00",
"deferDate": "2024-12-01T09:00:00",
"plannedDate": "2024-12-15T09:00:00",
"flagged": true,
"estimatedMinutes": 60,
"tagNames": ["Review", "Important"]
}
Planned Date vs Due Date:
dueDate: When the task must be completed (deadline)plannedDate: When you intend to work on the task (planning)- This distinction is crucial for separating deadlines from scheduled work time
omnifocus_complete_task
Mark a task as complete or dropped. You can identify the task by either ID or name.
{
"taskId": "abc123",
"action": "complete"
}
Or using task name:
{
"taskName": "Write documentation",
"action": "complete"
}
Action can be "complete" (default) or "drop". If both taskId and taskName are provided, taskId takes priority.
omnifocus_add_tag_to_task
Add a tag to a task. You can identify the task by either ID or name.
{
"taskId": "abc123",
"tagName": "Urgent"
}
Or using task name:
{
"taskName": "Write report",
"tagName": "Urgent"
}
If both taskId and taskName are provided, taskId takes priority.
omnifocus_remove_tag_from_task
Remove a tag from a task. You can identify the task by either ID or name.
{
"taskId": "abc123",
"tagName": "Urgent"
}
Or using task name:
{
"taskName": "Old task",
"tagName": "Done"
}
If both taskId and taskName are provided, taskId takes priority.
omnifocus_search
Search across OmniFocus.
{
"query": "report",
"searchType": "all",
"limit": 20
}
omnifocus_get_due_tasks
Get tasks due within a timeframe.
{
"daysAhead": 7,
"includeOverdue": true,
"limit": 50
}
omnifocus_get_flagged_tasks
Get flagged tasks.
{
"includeCompleted": false,
"limit": 50
}
omnifocus_get_planned_tasks
Get tasks planned within a timeframe.
{
"daysAhead": 7,
"includeOverdue": true,
"limit": 50
}
omnifocus_get_projects_for_review
Get projects that need review based on their next review date. Perfect for GTD practitioners following the review workflow.
{
"daysAhead": 0,
"status": "active",
"limit": 50
}
Parameters:
daysAhead: How many days ahead to look (0 = overdue reviews only)status: Filter by project status ("active", "done", "dropped", "onHold", "all")limit: Maximum number of projects to return (1-500)
omnifocus_mark_project_reviewed
Mark a project as reviewed and update its next review date. You can identify the project by either ID or name.
{
"projectId": "abc123"
}
Or using project name:
{
"projectName": "Weekly Review"
}
With custom review interval:
{
"projectName": "Work Project",
"reviewIntervalDays": 14
}
Parameters:
projectIdorprojectName: Identifies the project (ID takes priority)reviewIntervalDays(optional): Custom review interval in days. If not provided, uses the project's existing review interval.
omnifocus_batch_mark_reviewed
Mark multiple projects as reviewed in one efficient operation.
{
"projectIds": ["id1", "id2", "id3"]
}
With custom review interval for all:
{
"projectIds": ["id1", "id2", "id3"],
"reviewIntervalDays": 7
}
Parameters:
projectIds: Array of project IDs to mark as reviewed (1-100 projects)reviewIntervalDays(optional): Custom review interval to apply to all projects
Returns a summary with:
- Count of successful reviews
- Count of failures
- Full project data for successful reviews
- Error details for any failures
Date Formats
All dates use ISO 8601 format: YYYY-MM-DDTHH:mm:ss
Examples:
2024-12-31T17:00:00- December 31, 2024 at 5:00 PM2024-06-15T09:00:00- June 15, 2024 at 9:00 AM
Error Handling
The server provides clear error messages for common issues:
- OmniFocus not running: Launch OmniFocus first
- Permission denied: Enable automation permissions in System Preferences
- Item not found: The specified ID doesn't exist
- Invalid parameters: Check parameter format and values
Development
Build
npm run build
Watch mode
npm run dev
Test manually
After building, you can test with:
echo '{"jsonrpc":"2.0","method":"tools/list","id":1}' | node dist/index.js
License
MIT
Credits
Built using:
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.