Omnifocus

Provides an MCP server for OmniFocus to manage tasks, projects, folders, perspectives, and AI-assisted workflows.
  • javascript

0

GitHub Stars

javascript

Language

6 months ago

First Indexed

3 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs
{
  "mcpServers": {
    "mojenmojen-of-mcp": {
      "command": "node",
      "args": [
        "/path/to/of-mcp/dist/server.js"
      ]
    }
  }
}

You run an MCP server that connects OmniFocus to automated workflows and AI-assisted planning. This server exposes a rich set of task, project, and perspective management capabilities, enabling you to automate reviews, plan work, handle repeating tasks, organize with folders and perspectives, and integrate Claude AI for smarter workflows.

How to use

You operate the MCP server from your environment and interact with it through an MCP client that speaks the MCP protocol. Start the server, connect your client, then perform high-level actions like creating tasks and projects, organizing folders, and applying native OmniFocus perspectives. Use the server to drive project reviews, plan work with planned dates, manage repeating tasks, and access custom perspectives directly from OmniFocus. When you issue commands, rely on the server to return structured task data, IDs for deep linking, and batch results that let you perform bulk operations efficiently.

Key usage patterns include: batching write operations to create or update many items at once, filtering tasks across multiple projects with advanced criteria, and using native custom perspectives to display hierarchical task structures that reflect parent-child relationships. You can also leverage AI features to enhance tool descriptions and reduce confusion between perspectives and tags.

How to install

Prerequisites you need before installing the MCP server: Node.js 18+ and a macOS environment for OmniFocus usage. You should also have Claude Code available to enable Claude AI integration.

Quick install (recommended) then configure your MCP client to connect to the local server or a remote endpoint if you provide one.

# One-line installation
claude mcp add omnifocus-enhanced -- npx -y of-mcp

Alternative installation methods provide both a global install path and a local project setup. Use these if you prefer a global MCP client or want to host the server from a local project.

# Global installation
npm install -g of-mcp
claude mcp add omnifocus-enhanced -- of-mcp

# Local project installation
git clone https://github.com/mojenmojen/of-mcp.git
cd of-mcp
npm install && npm run build
claude mcp add omnifocus-enhanced -- node "/path/to/of-mcp/dist/server.js"

Configuration and runtime

The MCP server can be run locally via a standard Node command, typically pointing to the built server file. Use the exact runtime path shown in your installation steps to start the server in your environment.

{
  "mcpServers": {
    "omnifocus_mcp": {
      "command": "node",
      "args": ["/path/to/of-mcp/dist/server.js"]
    }
  }
}

Security and prerequisites

Ensure OmniFocus 3+ is installed and running, Node.js 18+ is available on your system, and Claude Code is configured for AI-assisted workflows. If you encounter permission prompts, grant accessibility permissions for terminal apps as needed.

Troubleshooting

If the MCP server does not start or cannot connect, verify the following: the server process is running, the command path is correct, the client is configured to point at the correct URL or local process, and Claude Code integration is enabled. Use a health check command to confirm connectivity and version.

Use cases

Project management and GTD-style workflows: create tasks and folders, manage subtasks, access built-in and custom perspectives, and apply batch operations for team productivity. AI-assisted planning with Claude helps prioritize and organize tasks based on your workflow.

Performance and reliability

The server supports true batching for write operations, fast ID lookups, lazy loading of OmniFocus collections, and efficient filtering. It is designed to scale with large task databases while keeping memory usage low.

Notes and further reading

You can leverage the server to enable automated weekly reviews, plan future tasks with planned dates, and explore deep interactions with custom perspectives inside OmniFocus for richer workflows.

Available tools

add_omnifocus_task

Create tasks (enhanced with subtask support) inside a specific project or by using IDs. Supports nested subtasks and flags for priority.

add_project

Create new projects within a folder or at the root level, with optional tag associations and metadata.

add_folder

Create folders and nest them, forming a hierarchical project structure.

remove_item

Delete a task or a project by ID or name, with safety checks to prevent accidental data loss.

edit_item

Edit attributes of tasks or projects, including due dates, status, project/folder assignments, and repetition rules.

batch_add_items

Add multiple tasks or projects in a single coordinated operation to speed up setup and templates.

batch_edit_items

Batch update multiple items with a single call, enabling bulk status changes, dates, and associations.

batch_remove_items

Remove multiple items in one operation to streamline cleanup.

get_task_by_id

Query task information by ID for quick lookups and deep links.

list_projects

List projects with optional folder and status filtering to navigate the workspace.

get_inbox_tasks

Retrieve tasks in the Inbox perspective with optional filters.

get_flagged_tasks

Fetch tasks marked for attention, with optional project-based filtering.

get_forecast_tasks

Retrieve upcoming tasks based on forecast parameters and due dates.

get_tasks_by_tag

Filter tasks by tag name or ID with options for multiple tags and matching modes.

list_tags

List all tags with optional counts and hierarchy to understand tagging structure.

filter_tasks

Advanced, multi-criteria task filtering for complex views and perspectives.

batch_filter_tasks

Batch filter tasks across multiple projects for faster multi-project queries.

list_custom_perspectives

List your OmniFocus custom perspectives and details for integration.

get_custom_perspective_tasks

Access tasks from a custom perspective with optional hierarchical display.

get_today_completed_tasks

View tasks completed today for quick productivity tracking.

get_projects_for_review

Fetch projects due for review to automate weekly or custom review workflows.

batch_mark_reviewed

Mark multiple projects as reviewed in a single operation.

get_server_version

Query the MCP server version and build information for diagnostics.

get_project_by_id

Query project information by ID or name, including review data.

get_folder_by_id

Query folder information by ID or name.

diagnose_connection

Check connectivity and permissions between OmniFocus and the MCP server.

search_tasks

Full-text search across task names and notes to locate items quickly.

duplicate_project

Create copies of projects, including tasks, with optional date shifting and folder placement.

edit_tag

Edit tag properties: rename, change status, move to parent, or set allowsNextAction.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational