← 返回
AI智能 中文

OmniFocus

Manage OmniFocus tasks via JavaScript for Automation (JXA) scripts. Use when the user asks Clawdbot to interact with OmniFocus, including - (1) Adding tasks to inbox, (2) Listing or searching tasks (inbox, available, flagged, overdue, due soon), (3) Completing tasks, (4) Updating task properties (notes, due dates, flags), (5) Getting OmniFocus statistics, (6) Reporting on task status, or (7) Acting on tasks in OmniFocus based on user queries.
通过JXA脚本管理OmniFocus任务。用于用户请求Clawdbot与OmniFocus交互时,包括:(1)添加任务到收件箱,(2)列出或搜索任务(收件箱/可执行/已标记/已逾期/即将到期),(3)完成任务,(4)更新任务属性(备注、到期日、标记),(5)获取统计信息,(6)报告任务状态,(7)根据查询执行任务操作。
shenzo1
AI智能 clawhub v1.0.0 1 版本 99771.3 Key: 无需
★ 2
Stars
📥 2,578
下载
💾 301
安装
1
版本
#latest

概述

OmniFocus Task Management

Automate OmniFocus task management via JavaScript for Automation (JXA) scripts.

Quick Start

All scripts are located in the scripts/ directory and use JXA. Run with:

osascript -l JavaScript scripts/<script-name>.js [args]

Key Scripts:

  • add_task.js - Add task to inbox
  • list_tasks.js - List tasks with filters
  • search_tasks.js - Search tasks by keyword
  • complete_task.js - Complete a task by name
  • update_task.js - Update task properties
  • get_stats.js - Get OmniFocus statistics

Core Operations

Adding Tasks

osascript -l JavaScript scripts/add_task.js "Task name" ["Note"] ["YYYY-MM-DD"]

Examples:

osascript -l JavaScript scripts/add_task.js "Buy groceries"
osascript -l JavaScript scripts/add_task.js "Review doc" "Check sections 1-3"
osascript -l JavaScript scripts/add_task.js "Submit report" "Q1" "2026-01-31"

Returns: Task ID

Listing Tasks

osascript -l JavaScript scripts/list_tasks.js [filter] [limit]

Filters:

  • inbox - Inbox tasks
  • available - Available (unblocked) tasks (default)
  • flagged - Flagged tasks
  • due-soon - Due within 3 days
  • overdue - Past due
  • all - All incomplete tasks

Returns: JSON array with task details (name, id, note, dueDate, flagged, project, tags)

Searching Tasks

osascript -l JavaScript scripts/search_tasks.js "keyword" [limit]

Searches task names and notes. Case-insensitive.

Returns: JSON array of matching tasks

Completing Tasks

osascript -l JavaScript scripts/complete_task.js "Task name"

Searches inbox first, then all tasks. Completes first match.

Updating Tasks

osascript -l JavaScript scripts/update_task.js "Task name" [--note "text"] [--due "YYYY-MM-DD"] [--flag true/false]

Examples:

osascript -l JavaScript scripts/update_task.js "Review" --note "Added notes"
osascript -l JavaScript scripts/update_task.js "Submit" --due "2026-02-01"
osascript -l JavaScript scripts/update_task.js "Important" --flag true

Getting Statistics

osascript -l JavaScript scripts/get_stats.js

Returns: JSON with counts:

  • total, incomplete, inbox
  • flagged, overdue, dueSoon
  • available, blocked

Usage Guidelines

When Responding to User Queries

  1. List tasks before acting on them to confirm targets
  2. Parse JSON output for structured processing
  3. Present results in user-friendly format (not raw JSON)
  4. Confirm operations before completing or modifying tasks
  5. Handle errors gracefully (task not found, etc.)

Common Patterns

Daily Review:

# Statistics overview
osascript -l JavaScript scripts/get_stats.js

# What needs attention
osascript -l JavaScript scripts/list_tasks.js overdue
osascript -l JavaScript scripts/list_tasks.js due-soon

Task Queries:

# "What's in my inbox?"
osascript -l JavaScript scripts/list_tasks.js inbox

# "What are my next actions?"
osascript -l JavaScript scripts/list_tasks.js available 10

# "Show my flagged tasks"
osascript -l JavaScript scripts/list_tasks.js flagged

Task Management:

# "Add a task to call John"
osascript -l JavaScript scripts/add_task.js "Call John"

# "Find tasks about the project"
osascript -l JavaScript scripts/search_tasks.js "project"

# "Mark 'Buy milk' as complete"
osascript -l JavaScript scripts/complete_task.js "Buy milk"

# "Flag the review task"
osascript -l JavaScript scripts/update_task.js "Review" --flag true

Output Handling

Scripts return JSON for structured data. When presenting to users:

  1. Parse the JSON
  2. Format results clearly
  3. Summarize counts and key information
  4. Highlight urgent items (overdue, due soon)

Example response format:

Found 3 overdue tasks:
• Submit Q1 report (due Jan 20)
• Review contract (due Jan 23)
• Call vendor (due Jan 24)

And 5 tasks due in the next 3 days:
• [list tasks]

Would you like me to flag or update any of these?

Error Handling

Common errors:

  • Task not found - Double-check name or search first
  • No tasks - Empty result, report clearly
  • Invalid date - Use YYYY-MM-DD format
  • OmniFocus not running - Scripts require OmniFocus

Multi-Step Operations

Find then act:

# 1. Search for task
RESULTS=$(osascript -l JavaScript scripts/search_tasks.js "meeting")

# 2. Parse and identify target task name

# 3. Complete the task
osascript -l JavaScript scripts/complete_task.js "Team meeting notes"

Technical Reference

For detailed API information and advanced usage, see:

  • JXA API Reference: references/jxa-api.md - Object model and methods
  • Automation Guide: references/automation-guide.md - Detailed script documentation and workflows

Read these files when:

  • Building complex queries
  • Understanding OmniFocus data model
  • Implementing custom workflows
  • Debugging scripts

Requirements

  • macOS
  • OmniFocus installed and running
  • Scripts have execute permissions (chmod +x)

Notes

  • Scripts use JXA (JavaScript for Automation), not AppleScript
  • Task matching is case-sensitive for exact names, case-insensitive for searches
  • Date format: YYYY-MM-DD (ISO 8601)
  • All operations are performed on the default OmniFocus document
  • Scripts are read-only safe except for add, complete, and update operations

版本历史

共 1 个版本

  • v1.0.0 当前
    2026-03-28 13:47 安全 安全

安全检测

腾讯云安全 (Keen)

安全,无风险
查看报告

腾讯云安全 (Sanbu)

安全,无风险
查看报告

🔗 相关推荐

ai-intelligence

ontology

oswalpalash
类型化知识图谱,用于结构化智能体记忆与可组合技能。支持创建/查询实体(人员、项目、任务、事件、文档)及关联...
★ 710 📥 243,653
ai-intelligence

self-improving agent

pskoett
捕获经验教训、错误和纠正,以实现持续改进。使用时机:(1)命令或操作意外失败;(2)用户纠正……
★ 4,058 📥 797,615
ai-intelligence

Proactive Agent

halthelobster
将AI智能体从任务执行者升级为主动预判需求、持续优化的智能伙伴。集成WAL协议、工作缓冲区、自主定时任务及实战验证模式。Hal Stack核心组件 🦞
★ 834 📥 212,935