Sponsored by Deepsite.site

Omnifocus

Created By
alexhannaonlinea day ago
An MCP server that gives AI assistants full access to OmniFocus via Omni Automation. 41 tools for tasks, projects, folders, tags, subtask hierarchies, notifications, attachments, analytics, and data export.
Overview

OmniFocus MCP Server

An MCP server that gives AI assistants full access to OmniFocus via Omni Automation. 41 tools for tasks, projects, folders, tags, subtask hierarchies, notifications, attachments, analytics, and data export.

How it works

The server runs osascript -l JavaScript to call app.evaluateJavascript(), which executes scripts inside OmniFocus's native Omni Automation engine. No direct database access — all interaction goes through the official API.

Installation

git clone https://github.com/alexhanna/omnifocus-mcp.git
cd omnifocus-mcp
npm install
npm run build

MCP client configuration

Add the server to your MCP client's configuration. The server entry looks like:

{
  "mcpServers": {
    "omnifocus": {
      "command": "node",
      "args": ["/path/to/omnifocus-mcp/dist/index.js"],
      "env": {
        "LOG_LEVEL": "info"
      }
    }
  }
}

Where this lives depends on your client:

  • Claude Desktop: ~/Library/Application Support/Claude/claude_desktop_config.json
  • Claude Code: ~/.claude/settings.json or project .mcp.json
  • Other MCP clients: consult your client's documentation for server configuration

Requirements

  • macOS (required for osascript)
  • OmniFocus 4.6+ installed
  • Node.js 18+
  • macOS automation permission for OmniFocus (granted on first use — see docs/PERMISSIONS.md)

Tools

Tasks (7 tools)

ToolDescription
list_tasksFilter by completion, flags, project, tags, dates, search, inbox, availability. Cached 30s.
get_task_countCount matching tasks without returning data.
todays_agendaTasks due today, overdue, or flagged.
create_taskCreate task with name, note, flags, dates, tags, project, estimated time.
update_taskModify any task property. Move between projects.
complete_taskMark task completed.
delete_taskRemove task.

Projects (5 tools)

ToolDescription
list_projectsFilter by status, flags, folder. Cached 5min.
create_projectCreate with folder placement (auto-creates folder if needed).
update_projectChange properties, move between folders.
complete_projectMark project done (optionally complete all tasks).
delete_projectRemove project.

Folders (6 tools)

ToolDescription
list_foldersList with nesting depth and status filter.
create_folderCreate folder, optionally nested.
update_folderRename, change status.
delete_folderDrop folder (optionally move contents first).
get_folder_contentsList projects and subfolders in a folder.
move_to_folderMove project or folder into a different folder.

Task Hierarchies (5 tools)

ToolDescription
get_task_hierarchyGet task with full subtask tree (configurable depth).
create_subtaskCreate child task with position control.
move_taskMove task to different parent or project root.
set_task_orderingSet sequential/parallel and completedByChildren.
list_subtasksList direct children of a task.

Tags (2 tools)

ToolDescription
list_tagsList all tags. Optional (expensive) usage statistics.
manage_tagsCreate, rename, delete, merge tags. Supports nested tags via parentTag.

Notifications (4 tools)

ToolDescription
list_notificationsList all notifications on a task.
add_notificationAdd absolute (date) or relative (offset from due) notification.
remove_notificationRemove notification by index.
clear_notificationsRemove all notifications from a task.

Attachments (4 tools)

ToolDescription
list_attachmentsList attachment metadata and linked files on a task.
add_linked_fileLink a file:// path to a task.
remove_attachmentRemove embedded attachment by index.
remove_linked_fileUnlink a file reference.

Analytics (3 tools)

ToolDescription
get_productivity_statsCompletion rates, time distribution. Group by project/tag/day/week.
get_task_velocityTask throughput over day/week/month.
analyze_overdue_tasksOverdue patterns grouped by project/tag/age/priority.

Recurring Tasks (2 tools)

ToolDescription
analyze_recurring_tasksList recurring tasks with frequency and due dates.
get_recurring_patternsStatistics about recurring task frequencies.

Export (3 tools)

ToolDescription
export_tasksExport tasks as JSON or CSV with field selection.
export_projectsExport projects with optional statistics.
bulk_exportExport all data (tasks, projects, tags) to files.

Architecture

src/
├── cache/           # TTL-based caching (30s tasks, 5min projects, 1h analytics)
├── omnifocus/
│   ├── OmniAutomation.ts   # Bridge: osascript → evaluateJavascript()
│   ├── scripts/             # Omni Automation script templates
│   └── types.ts             # Type definitions
├── tools/           # 41 MCP tools in 10 domain directories
│   ├── tasks/       ├── projects/    ├── folders/
│   ├── hierarchy/   ├── tags/        ├── notifications/
│   ├── attachments/ ├── analytics/   ├── recurring/
│   └── export/
├── utils/           # Logging, permission checking
└── index.ts         # Server entry point

Script execution flow:

  1. Tool calls omniAutomation.buildScript(template, params) — replaces {{placeholders}} with JSON.stringify-escaped values
  2. wrapScript() wraps the script in a try/catch IIFE
  3. The wrapped script is JSON.stringify-escaped again and passed to app.evaluateJavascript()
  4. osascript -l JavaScript runs the outer JXA wrapper
  5. JSON result is parsed and returned through MCP

Known limitations

  • No recurring task creation — can read recurring patterns but cannot create repeat rules
  • No perspective management — Omni Automation API is read-only for perspectives
  • No location-based tags — must be set manually in OmniFocus UI
  • add_linked_file only accepts file:// URLs — OmniFocus rejects http/https
  • MCP server caches builds — Claude Code must restart to pick up npm run build changes
  • No review/perspective management via automation

Development

npm run build     # Compile TypeScript
npm run dev       # Watch mode
npm test          # Unit tests (vitest)
npm run typecheck # Type check without building

License

MIT — see LICENSE

Server Config

{
  "mcpServers": {
    "omnifocus": {
      "command": "node",
      "args": [
        "/path/to/omnifocus-mcp/dist/index.js"
      ],
      "env": {
        "LOG_LEVEL": "info"
      }
    }
  }
}
Recommend Servers
TraeBuild with Free GPT-4.1 & Claude 3.7. Fully MCP-Ready.
Serper MCP ServerA Serper MCP Server
DeepChatYour AI Partner on Desktop
BlenderBlenderMCP connects Blender to Claude AI through the Model Context Protocol (MCP), allowing Claude to directly interact with and control Blender. This integration enables prompt assisted 3D modeling, scene creation, and manipulation.
EdgeOne Pages MCPAn MCP service designed for deploying HTML content to EdgeOne Pages and obtaining an accessible public URL.
Y GuiA web-based graphical interface for AI chat interactions with support for multiple AI models and MCP (Model Context Protocol) servers.
Amap Maps高德地图官方 MCP Server
ChatWiseThe second fastest AI chatbot™
TimeA Model Context Protocol server that provides time and timezone conversion capabilities. This server enables LLMs to get current time information and perform timezone conversions using IANA timezone names, with automatic system timezone detection.
AiimagemultistyleA Model Context Protocol (MCP) server for image generation and manipulation using fal.ai's Stable Diffusion model.
MiniMax MCPOfficial MiniMax Model Context Protocol (MCP) server that enables interaction with powerful Text to Speech, image generation and video generation APIs.
Baidu Map百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
CursorThe AI Code Editor
Jina AI MCP ToolsA Model Context Protocol (MCP) server that integrates with Jina AI Search Foundation APIs.
MCP AdvisorMCP Advisor & Installation - Use the right MCP server for your needs
Tavily Mcp
Zhipu Web SearchZhipu Web Search MCP Server is a search engine specifically designed for large models. It integrates four search engines, allowing users to flexibly compare and switch between them. Building upon the web crawling and ranking capabilities of traditional search engines, it enhances intent recognition capabilities, returning results more suitable for large model processing (such as webpage titles, URLs, summaries, site names, site icons, etc.). This helps AI applications achieve "dynamic knowledge acquisition" and "precise scenario adaptation" capabilities.
Playwright McpPlaywright MCP server
Howtocook Mcp基于Anduin2017 / HowToCook (程序员在家做饭指南)的mcp server,帮你推荐菜谱、规划膳食,解决“今天吃什么“的世纪难题; Based on Anduin2017/HowToCook (Programmer's Guide to Cooking at Home), MCP Server helps you recommend recipes, plan meals, and solve the century old problem of "what to eat today"
WindsurfThe new purpose-built IDE to harness magic
Visual Studio Code - Open Source ("Code - OSS")Visual Studio Code