omnifocus-mcp-enhanced 1.6.9 → 1.6.11

package/.github/workflows/ci.yml

@@ -0,0 +1,21 @@
+name: CI
+
+on:
+  push:
+    branches: [main]
+  pull_request:
+
+permissions:
+  contents: read
+
+jobs:
+  test:
+    runs-on: macos-latest
+    steps:
+      - uses: actions/checkout@v4
+      - uses: actions/setup-node@v4
+        with:
+          node-version: 20
+          cache: npm
+      - run: npm ci
+      - run: npm test

package/LICENSE.md

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2025-2026 jqlts1
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

package/README.md

@@ -13,10 +13,38 @@
   <img width="380" height="200" src="https://glama.ai/mcp/servers/@jqlts1/omnifocus-mcp-enhanced/badge" alt="OmniFocus Enhanced MCP server" />
 </a>
 
-Enhanced Model Context Protocol (MCP) server for OmniFocus featuring **native custom perspective access**, hierarchical task display, AI-optimized tool selection, and comprehensive task management. Perfect integration with Claude AI for intelligent workflows.
+Enhanced Model Context Protocol (MCP) server for OmniFocus featuring **native custom perspective access**, hierarchical task display, AI-optimized tool selection, and comprehensive task management.
+
+In plain English: this lets your AI assistant read your OmniFocus data, create tasks/projects, organize subtasks, review perspectives, and help you plan work without you manually jumping between apps.
+
+## 🌠 Why This Project Exists
+
+OmniFocus is already powerful, but it is still mostly a tool you drive by hand.
+
+The bigger idea behind this project is simple:
+
+- less clicking, more conversation
+- less manual cleanup, more AI-assisted planning
+- less tool memorization, more natural task management
+
+The goal is not just to expose more OmniFocus commands.
+The goal is to let you work with OmniFocus like this:
+
+```text
+Plan my day.
+Clean up my Inbox.
+Turn these notes into a project.
+Show me what is blocked.
+Reorganize these tasks safely.
+```
+
+If that feels natural, this MCP server is doing its job.
+
+Want to see where the project is heading next? See the [roadmap](docs/roadmap/2026-02-25-batch-move-tasks-plan.md).
 
 ## 🆕 Latest Release
 
+- **v1.6.10** - Fixed Inbox task completion via `edit_item`, fixed AppleScript special-character handling for apostrophes/backslashes, fixed JSON result escaping for special characters, and clarified `batch_add_items` / `mcporter` usage with working examples.
 - **v1.6.9** - Added task attachment support: `get_task_by_id` now lists attachment metadata, `dump_database` exports attachment/link metadata, and new `read_task_attachment` returns image attachments as MCP image content when possible.
 - **v1.6.8** - Added stable task move support via `move_task` and `edit_item` (`newProjectId/newProjectName/newParentTaskId/newParentTaskName/moveToInbox`) with duplicate-name protection and cycle-prevention checks.
 - **v1.6.6** - Added full Planned Date support (create/edit/read/filter/sort/export), including `plannedDate`/`newPlannedDate` and updated task displays.
@@ -43,14 +71,15 @@ Enhanced Model Context Protocol (MCP) server for OmniFocus featuring **native cu
 
 ## 📦 Installation
 
-### Quick Install (Recommended)
+### Claude Code
 
+#### Quick Install (recommended)
 ```bash
 # One-line installation
 claude mcp add omnifocus-enhanced -- npx -y omnifocus-mcp-enhanced
 ```
 
-### Alternative Installation Methods
+#### Alternative methods for Claude Code:
 
 ```bash
 # Upgrade to latest
@@ -67,13 +96,137 @@ npm install && npm run build
 claude mcp add omnifocus-enhanced -- node "/path/to/omnifocus-mcp-enhanced/dist/server.js"
 ```
 
+### Claude Desktop / Cowork
+
+Add the server to `~/Library/Application Support/Claude/claude_desktop_config.json`:
+
+```json
+{
+  "mcpServers": {
+    "omnifocus-enhanced": {
+      "command": "npx",
+      "args": ["-y", "omnifocus-mcp-enhanced"]
+    }
+  }
+}
+```
+
+For a local clone, use:
+
+```json
+{
+  "mcpServers": {
+    "omnifocus-enhanced": {
+      "command": "node",
+      "args": ["/path/to/omnifocus-mcp-enhanced/dist/server.js"]
+    }
+  }
+}
+```
+
+Restart Claude Desktop after editing the config file.
+
+### Using Both Claude Code and Claude Desktop / Cowork
+
+Claude Code and Claude Desktop read separate configurations. If you want to use this MCP server from both, you need to install it in both places — run `claude mcp add` for Claude Code **and** add the entry to `claude_desktop_config.json` for Claude Desktop / Cowork.
+
 ## 📋 Requirements
 
 - **macOS 10.15+** - OmniFocus is macOS-only
 - **OmniFocus 3+** - The application must be installed and running
 - **OmniFocus Pro** - Required for custom perspectives (new features in v1.6.0)
 - **Node.js 18+** - For running the MCP server
-- **Claude Code** - For MCP integration
+- **Any MCP-capable client** - Claude Code, `mcporter`, or another MCP host
+
+## 🚦 Start Here
+
+If you only want the fastest way to understand this project, remember this:
+
+1. Connect the MCP server to your AI client.
+2. Talk to the AI naturally.
+3. Let it read, plan, create, move, or update your OmniFocus tasks for you.
+
+You do not need to memorize all tool names first.
+
+## 🙋 What This Is Good For
+
+- **Daily planning**: ask your AI what is due today, what is flagged, and what you can finish in 30 minutes.
+- **Project setup**: give the AI a rough goal, then let it create a project and break it into subtasks.
+- **Inbox cleanup**: ask it to review Inbox tasks and sort them into next actions, projects, or someday/later buckets.
+- **Perspective reviews**: ask it to open one of your custom perspectives and summarize what matters.
+- **Batch capture**: paste meeting notes or a brainstorm list and let the AI create multiple tasks at once.
+- **Attachment-aware review**: let the AI inspect task attachments only when needed.
+
+## 💬 Example AI Conversations
+
+These work well in Claude Code or any MCP client that can call the same tools.
+
+### 1. Daily Planning
+
+Try saying:
+
+```text
+Check my Forecast and flagged tasks, then tell me the 3 most important things to do today.
+Prefer tasks that take under 60 minutes first.
+```
+
+### 2. Inbox Cleanup
+
+Try saying:
+
+```text
+Review my Inbox and group the tasks into:
+1. do today
+2. schedule later
+3. turn into projects
+Then help me clean up the obvious ones.
+```
+
+### 3. Turn an Idea Into a Project
+
+Try saying:
+
+```text
+Create a project called "Launch spring newsletter".
+Add the main subtasks, estimated minutes, and mark the most important step as flagged.
+```
+
+### 4. Use a Custom Perspective
+
+Try saying:
+
+```text
+Open my custom perspective "今日工作安排" and summarize:
+- what is due soon
+- what looks blocked
+- what I can finish quickly
+```
+
+### 5. Batch Add From Notes
+
+Try saying:
+
+```text
+Turn these meeting notes into OmniFocus tasks under the project "Website Refresh".
+Use subtasks where it makes sense and keep the task names short.
+```
+
+### 6. Review Attachments Only When Needed
+
+Try saying:
+
+```text
+Find the task called "Review design draft".
+Show me what attachments it has first.
+Only open the image attachment if there is one.
+```
+
+## 🧭 Practical Usage Tips
+
+- Ask the AI to **look first, then change things** if you want safer workflows.
+- Use **task IDs** when you have duplicate task names.
+- For **subtasks**, let the parent task determine the project. Do not also pass `projectName`.
+- For `mcporter`, complex arrays are much more reliable with `--args '{...}'`.
 
 ## 🎯 Core Capabilities
 
@@ -217,6 +370,101 @@ Efficiently manage multiple tasks:
 }
 ```
 
+CLI tip for `mcporter`:
+
+```bash
+# Prefer explicit JSON args for complex arrays / nested objects
+mcporter call omnifocus.batch_add_items --args '{
+  "items": [
+    {
+      "type": "task",
+      "name": "Website Technical SEO",
+      "projectName": "SEO Project"
+    }
+  ]
+}'
+```
+
+If you pass a subtask with `parentTaskId` or `parentTaskName`, do not also pass `projectName`. Subtasks inherit the project from their parent task.
+
+Working `mcporter` examples:
+
+```bash
+# 1) Batch-create top-level tasks in a project
+mcporter call omnifocus.batch_add_items --args '{
+  "items": [
+    {
+      "type": "task",
+      "name": "Parent: Category A",
+      "projectName": "OmniFocus MCP Batch Test"
+    },
+    {
+      "type": "task",
+      "name": "Parent: Category B",
+      "projectName": "OmniFocus MCP Batch Test"
+    }
+  ]
+}'
+```
+
+```bash
+# 2) Create parent + child in one batch
+mcporter call omnifocus.batch_add_items --args '{
+  "items": [
+    {
+      "type": "task",
+      "name": "Parent: Category A",
+      "projectName": "OmniFocus MCP Batch Test"
+    },
+    {
+      "type": "task",
+      "name": "Child: A1",
+      "parentTaskName": "Parent: Category A"
+    }
+  ]
+}'
+```
+
+```bash
+# 3) Safer two-step flow when adding many subtasks to existing parents
+mcporter call omnifocus.batch_add_items --args '{
+  "items": [
+    {
+      "type": "task",
+      "name": "Child: A1",
+      "parentTaskName": "Parent: Category A"
+    },
+    {
+      "type": "task",
+      "name": "Child: A2",
+      "parentTaskName": "Parent: Category A"
+    },
+    {
+      "type": "task",
+      "name": "Child: B1",
+      "parentTaskName": "Parent: Category B"
+    }
+  ]
+}'
+```
+
+This will fail, by design:
+
+```bash
+mcporter call omnifocus.batch_add_items --args '{
+  "items": [
+    {
+      "type": "task",
+      "name": "Child: A1",
+      "projectName": "OmniFocus MCP Batch Test",
+      "parentTaskName": "Parent: Category A"
+    }
+  ]
+}'
+```
+
+Because a subtask must inherit its project from the parent task.
+
 ### 6. 🖼️ Attachment Inspection
 
 Discover images and linked files on a task first, then read only the attachment you need:
@@ -374,7 +622,10 @@ get_custom_perspective_tasks {
 
 ## 🔧 Configuration
 
-### Verify Installation
+### Claude Code
+
+Verify the server is registered:
+
 ```bash
 # Check MCP status
 claude mcp list
@@ -386,10 +637,15 @@ get_inbox_tasks
 list_custom_perspectives
 ```
 
+### Claude Desktop / Cowork
+
+Open `~/Library/Application Support/Claude/claude_desktop_config.json` and confirm the `omnifocus-enhanced` entry is present under `mcpServers`. Restart the app after any changes. Once running, you can test by asking the assistant to list your inbox tasks or custom perspectives.
+
 ### Troubleshooting
 - Ensure OmniFocus 3+ is installed and running
 - Verify Node.js 18+ is installed
-- Check Claude Code MCP configuration
+- For Claude Code: run `claude mcp list` to confirm the server is registered
+- For Claude Desktop / Cowork: verify `claude_desktop_config.json` is valid JSON and restart the app
 - Enable accessibility permissions for terminal apps if needed
 
 ## 🎯 Use Cases

package/README.zh.md

@@ -9,10 +9,38 @@
 
 > **将 OmniFocus 转换为 AI 驱动的生产力强化工具，支持自定义透视**
 
-增强版 OmniFocus 模型上下文协议（MCP）服务器，具备**原生自定义透视访问**、层级任务显示、AI 优化工具选择和全面的任务管理功能。与 Claude AI 完美集成，实现智能工作流。
+增强版 OmniFocus 模型上下文协议（MCP）服务器，具备**原生自定义透视访问**、层级任务显示、AI 优化工具选择和全面的任务管理功能。
+
+说人话：它可以让你的 AI 助手直接读取 OmniFocus、创建任务和项目、拆子任务、查看透视、做每日规划，不需要你自己在 OmniFocus 里来回点很多次。
+
+## 🌠 这个项目为什么存在
+
+OmniFocus 本身已经很强了，但它大多数时候仍然是一个需要你手动操作的工具。
+
+这个项目想做的，其实很简单：
+
+- 少点很多次按钮，多用自然对话
+- 少做手工整理，多让 AI 帮你规划
+- 少记工具名，多直接说你想完成什么
+
+目标不只是继续往外暴露更多 OmniFocus 命令。
+更重要的是，让你以后可以这样使用 OmniFocus：
+
+```text
+帮我规划今天。
+帮我清理 Inbox。
+把这段笔记变成一个项目。
+告诉我哪些事情卡住了。
+把这批任务安全地重新整理一下。
+```
+
+如果 README 读到这里，你已经能感受到这个方向，那这段说明就有价值了。
+
+如果你想继续看这个项目接下来准备往哪里走，可以直接看[路线图](docs/roadmap/2026-02-25-batch-move-tasks-plan.zh.md)。
 
 ## 🆕 最新版本
 
+- **v1.6.10** - 修复 `edit_item` 无法完成 Inbox 任务的问题，修复 AppleScript 对单引号/反斜杠等特殊字符的处理，修复特殊字符导致的 JSON 返回解析失败，并补充 `batch_add_items` / `mcporter` 的可直接运行示例与说明。
 - **v1.6.9** - 新增任务附件支持：`get_task_by_id` 会返回附件元信息，`dump_database` 导出附件/链接元信息，并新增 `read_task_attachment`，可在支持时直接把图片附件作为 MCP 图片内容返回。
 - **v1.6.6** - 新增 Planned Date（计划日期）全链路支持：创建/编辑/读取/过滤/排序/导出，包含 `plannedDate` / `newPlannedDate`。
 
@@ -68,7 +96,97 @@ claude mcp add omnifocus-enhanced -- node "/path/to/omnifocus-mcp-enhanced/dist/
 - **OmniFocus 3+** - 必须安装并运行该应用程序
 - **OmniFocus Pro** - 自定义透视功能需要 Pro 版本（v1.6.0 新功能）
 - **Node.js 18+** - 运行 MCP 服务器
-- **Claude Code** - MCP 集成
+- **任意支持 MCP 的客户端** - 比如 Claude Code、`mcporter` 或其他 MCP Host
+
+## 🚦 先看这里
+
+如果你想最快理解这个项目，只要记住这 3 件事：
+
+1. 把这个 MCP server 接到你的 AI 客户端里。
+2. 直接用自然语言和 AI 说你要做什么。
+3. 让 AI 帮你读 OmniFocus、整理任务、创建项目、拆子任务、做规划。
+
+你一开始不需要背所有工具名。
+
+## 🙋 这个项目最适合拿来做什么
+
+- **每日规划**：让 AI 看今天到期、已标记、可快速完成的任务。
+- **项目拆解**：给 AI 一个目标，让它自动建项目并拆成子任务。
+- **Inbox 清理**：让 AI 帮你把收件箱分成“今天做 / 以后排 / 变项目”。
+- **透视复盘**：让 AI 打开你的自定义透视并做总结。
+- **批量录入**：把会议纪要、脑暴清单直接变成一批任务。
+- **按需看附件**：先看任务有哪些附件，再决定要不要让 AI 打开。
+
+## 💬 和大模型对话的示例
+
+下面这些说法，在 Claude Code 或其他支持 MCP 的客户端里都很适合直接用。
+
+### 1. 每日规划
+
+你可以直接说：
+
+```text
+看看我今天的 Forecast 和已标记任务，然后告诉我今天最重要的 3 件事。
+优先考虑 60 分钟以内能完成的任务。
+```
+
+### 2. 清理 Inbox
+
+你可以直接说：
+
+```text
+帮我看一下 Inbox，把这些任务分成：
+1. 今天做
+2. 以后安排
+3. 应该升级成项目
+然后顺手把明显的项整理掉。
+```
+
+### 3. 把一个想法变成项目
+
+你可以直接说：
+
+```text
+创建一个项目，名字叫“春季 newsletter 发布”。
+把主要步骤拆成子任务，补上预计时间，并把最关键的一步设成 flagged。
+```
+
+### 4. 使用自定义透视
+
+你可以直接说：
+
+```text
+打开我的自定义透视“今日工作安排”，帮我总结：
+- 哪些快到期
+- 哪些像是卡住了
+- 哪些可以快速做完
+```
+
+### 5. 根据笔记批量创建任务
+
+你可以直接说：
+
+```text
+把这段会议纪要整理成 OmniFocus 任务，放到“网站改版”项目下。
+该拆成子任务的就拆，任务名尽量简短。
+```
+
+### 6. 只在需要时查看附件
+
+你可以直接说：
+
+```text
+找到“检查设计稿”这个任务。
+先告诉我它有哪些附件。
+如果里面有图片，再帮我打开图片附件。
+```
+
+## 🧭 实用建议
+
+- 如果你想更稳一点，可以先让 AI **先查看，再修改**。
+- 如果有重名任务，优先用 **task ID**。
+- 创建**子任务**时，让父任务决定项目，不要再额外传 `projectName`。
+- 在 `mcporter` 里，复杂数组参数尽量用 `--args '{...}'`。
 
 ## 🎯 核心功能
 
@@ -212,6 +330,101 @@ get_custom_perspective_tasks {
 }
 ```
 
+`mcporter` 调用提示：
+
+```bash
+# 复杂数组 / 嵌套对象，建议明确使用 --args JSON
+mcporter call omnifocus.batch_add_items --args '{
+  "items": [
+    {
+      "type": "task",
+      "name": "网站技术 SEO",
+      "projectName": "SEO 项目"
+    }
+  ]
+}'
+```
+
+如果某条子任务传了 `parentTaskId` 或 `parentTaskName`，就不要再传 `projectName`。子任务会自动继承父任务所在项目。
+
+可直接运行的 `mcporter` 示例：
+
+```bash
+# 1）批量创建项目下的顶层任务
+mcporter call omnifocus.batch_add_items --args '{
+  "items": [
+    {
+      "type": "task",
+      "name": "父任务：分类A",
+      "projectName": "OmniFocus MCP 批量测试"
+    },
+    {
+      "type": "task",
+      "name": "父任务：分类B",
+      "projectName": "OmniFocus MCP 批量测试"
+    }
+  ]
+}'
+```
+
+```bash
+# 2）单次批量里同时创建父任务和子任务
+mcporter call omnifocus.batch_add_items --args '{
+  "items": [
+    {
+      "type": "task",
+      "name": "父任务：分类A",
+      "projectName": "OmniFocus MCP 批量测试"
+    },
+    {
+      "type": "task",
+      "name": "子任务：A1",
+      "parentTaskName": "父任务：分类A"
+    }
+  ]
+}'
+```
+
+```bash
+# 3）更稳妥的两步法：父任务已存在时，再批量创建多个子任务
+mcporter call omnifocus.batch_add_items --args '{
+  "items": [
+    {
+      "type": "task",
+      "name": "子任务：A1",
+      "parentTaskName": "父任务：分类A"
+    },
+    {
+      "type": "task",
+      "name": "子任务：A2",
+      "parentTaskName": "父任务：分类A"
+    },
+    {
+      "type": "task",
+      "name": "子任务：B1",
+      "parentTaskName": "父任务：分类B"
+    }
+  ]
+}'
+```
+
+下面这种写法会失败，这属于预期行为：
+
+```bash
+mcporter call omnifocus.batch_add_items --args '{
+  "items": [
+    {
+      "type": "task",
+      "name": "子任务：A1",
+      "projectName": "OmniFocus MCP 批量测试",
+      "parentTaskName": "父任务：分类A"
+    }
+  ]
+}'
+```
+
+因为子任务必须继承父任务所在项目，不能再单独传 `projectName`。
+
 ### 6. 🖼️ 附件查看
 
 先读取任务和附件元信息，再按需打开具体附件：

package/dist/server.js

@@ -18,6 +18,7 @@ import * as getInboxTasksTool from './tools/definitions/getInboxTasks.js';
 import * as getFlaggedTasksTool from './tools/definitions/getFlaggedTasks.js';
 import * as getForecastTasksTool from './tools/definitions/getForecastTasks.js';
 import * as getTasksByTagTool from './tools/definitions/getTasksByTag.js';
+import * as listTagsTool from './tools/definitions/listTags.js';
 // Import ultimate filter tool
 import * as filterTasksTool from './tools/definitions/filterTasks.js';
 // Import custom perspective tools
@@ -45,6 +46,7 @@ server.tool("get_inbox_tasks", "Get tasks from OmniFocus inbox perspective", get
 server.tool("get_flagged_tasks", "Get flagged tasks from OmniFocus with optional project filtering", getFlaggedTasksTool.schema.shape, getFlaggedTasksTool.handler);
 server.tool("get_forecast_tasks", "Get tasks from OmniFocus forecast perspective (due/deferred tasks in date range)", getForecastTasksTool.schema.shape, getForecastTasksTool.handler);
 server.tool("get_tasks_by_tag", "Get tasks filtered by OmniFocus tags (labels like @home, @work, @urgent). Use this for tag-based filtering, NOT for custom perspective names. Tags are labels assigned to individual tasks.", getTasksByTagTool.schema.shape, getTasksByTagTool.handler);
+server.tool("list_tags", "List OmniFocus tags with IDs, parent relationships, and active status without loading tasks", listTagsTool.schema.shape, listTagsTool.handler);
 // Ultimate filter tool - The most powerful task perspective engine
 server.tool("filter_tasks", "Advanced task filtering with unlimited perspective combinations - status, dates, projects, tags, search, and more", filterTasksTool.schema.shape, filterTasksTool.handler);
 // Custom perspective tools

package/dist/tools/definitions/addOmniFocusTask.js

@@ -42,7 +42,7 @@ export async function handler(args, extra) {
             return {
                 content: [{
                         type: "text",
-                        text: `✅ Task "${args.name}" created successfully ${locationText}${dueDateText}${plannedDateText}${tagText}.`
+                        text: `✅ Task "${args.name}" created successfully ${locationText}${dueDateText}${plannedDateText}${tagText}.\n\nid: ${result.taskId}`
                     }]
             };
         }

package/dist/tools/definitions/addProject.js

@@ -36,7 +36,7 @@ export async function handler(args, extra) {
             return {
                 content: [{
                         type: "text",
-                        text: `✅ Project "${args.name}" created successfully ${locationText}${dueDateText}${plannedDateText}${tagText}${sequentialText}.`
+                        text: `✅ Project "${args.name}" created successfully ${locationText}${dueDateText}${plannedDateText}${tagText}${sequentialText}.\n\nid: ${result.projectId}`
                     }]
             };
         }

package/dist/tools/definitions/batchAddItems.js

@@ -12,9 +12,9 @@ export const schema = z.object({
         estimatedMinutes: z.number().optional().describe("Estimated time to complete the item, in minutes"),
         tags: z.array(z.string()).optional().describe("Tags to assign to the item"),
         // Task-specific properties
-        projectName: z.string().optional().describe("For tasks: The name of the project to add the task to"),
-        parentTaskId: z.string().optional().describe("For tasks: The ID of the parent task to create this task as a subtask"),
-        parentTaskName: z.string().optional().describe("For tasks: The name of the parent task to create this task as a subtask"),
+        projectName: z.string().optional().describe("For tasks: The project name for top-level tasks. Omit this when parentTaskId or parentTaskName is set."),
+        parentTaskId: z.string().optional().describe("For tasks: The parent task ID for subtasks. When this is set, do not also provide projectName."),
+        parentTaskName: z.string().optional().describe("For tasks: The parent task name for subtasks. Subtasks inherit project from their parent, so do not also provide projectName."),
         // Project-specific properties
         folderName: z.string().optional().describe("For projects: The name of the folder to add the project to"),
         sequential: z.boolean().optional().describe("For projects: Whether tasks in the project should be sequential")
@@ -36,7 +36,7 @@ export async function handler(args, extra) {
                 if (item.success) {
                     const itemType = args.items[index].type;
                     const itemName = args.items[index].name;
-                    return `- ✅ ${itemType}: "${itemName}"`;
+                    return `- ✅ ${itemType}: "${itemName}" (id: ${item.id})`;
                 }
                 else {
                     const itemType = args.items[index].type;