lark-task

# task +assign

> **Prerequisites:** Please read `../lark-shared/SKILL.md` to understand authentication, global parameters, and security rules.

Assign or remove members (assignees) from a task.

## Recommended Commands

```bash
# Add an assignee
lark-cli task +assign --task-id "<task_guid>" --add "ou_aaa"

# Add an app assignee
lark-cli task +assign --task-id "<task_guid>" --add "cli_xxx"

# Transfer an assignee (remove old, add new)
lark-cli task +assign --task-id "<task_guid>" --remove "ou_old" --add "ou_new"

# Add multiple assignees
lark-cli task +assign --task-id "<task_guid>" --add "ou_aaa,ou_bbb"
```

## Parameters

| Parameter | Required | Description |
|-----------|----------|-------------|
| `--task-id <guid>` | Yes | The task GUID to modify. For Feishu task applinks, use the `guid` query parameter, not the `suite_entity_num` / display task ID like `t104121`. |
| `--add <ids>` | No | Comma-separated assignee IDs. Use user `open_id`s like `ou_xxx` for people, or app IDs like `cli_xxx` for apps. |
| `--remove <ids>` | No | Comma-separated assignee IDs. Use user `open_id`s like `ou_xxx` for people, or app IDs like `cli_xxx` for apps. |

## Workflow

1. Confirm the task and members to add/remove.
2. Execute the command.
3. Report success and the new count of assignees.

> [!CAUTION]
> This is a **Write Operation** -- You must confirm the user's intent before executing.

# task +comment

> **Prerequisites:** Please read `../lark-shared/SKILL.md` to understand authentication, global parameters, and security rules.

Add a comment to an existing task.

## Recommended Commands

```bash
# Add a comment
lark-cli task +comment --task-id "<task_guid>" --content "Looks good!"
```

## Parameters

| Parameter | Required | Description |
|-----------|----------|-------------|
| `--task-id <guid>` | Yes | The task GUID to comment on. For Feishu task applinks, use the `guid` query parameter, not the `suite_entity_num` / display task ID like `t104121`. |
| `--content <text>` | Yes | The text content of the comment. |

## Workflow

1. Confirm the task and comment content.
2. Execute `lark-cli task +comment --task-id "..." --content "..."`
3. Report success and comment ID.

> [!CAUTION]
> This is a **Write Operation** -- You must confirm the user's intent before executing.

# task +complete

> **Prerequisites:** Please read `../lark-shared/SKILL.md` to understand authentication, global parameters, and security rules.

Mark a task as completed.

## Recommended Commands

```bash
# Complete a task
lark-cli task +complete --task-id "<task_guid>"

# A task applink is accepted directly; the CLI extracts its guid query value
lark-cli task +complete --task-id "https://applink.larksuite.com/client/todo/task?guid=<task_guid>"
```

## Parameters

| Parameter | Required | Description |
|-----------|----------|-------------|
| `--task-id <guid-or-applink>` | Yes | Task OpenAPI GUID or a task applink containing `guid=`. Display task IDs such as `t104121` / `suite_entity_num` are rejected. |

## Workflow

1. Confirm the task to complete.
2. Execute the command.
3. Read `data.status`, `data.completed_at`, and `data.already_completed` from the result. `already_completed: true` means the shortcut observed an already-completed task and skipped the PATCH.
4. Do not routinely call `task tasks get` when the result already reports `status: done` and a non-zero `completed_at`. Query details only if confirmation fields are absent or the user explicitly asks for a full verification.

> [!CAUTION]
> This is a **Write Operation** -- You must confirm the user's intent before executing.

# task +create

> **Prerequisites:** Please read `../lark-shared/SKILL.md` to understand authentication, global parameters, and security rules.

Create a new task in Lark.

## Recommended Commands

```bash
# Create a task with all details
lark-cli task +create \
  --summary "Quarterly Sales Review" \
  --description "Review the sales performance for the last quarter." \
  --assignee "ou_xxx" \
  --due "2026-03-25" \
  --tasklist-id "https://applink.larkoffice.com/client/todo/task_list?guid=a4b00000-000-000-000-00000000036c"

# Create a task assigned to an app
lark-cli task +create \
  --summary "Nightly Sync" \
  --assignee "cli_xxx"

# Create a simple task
lark-cli task +create \
  --summary "Buy milk"

# Create a milestone by passing an API field without a named flag
lark-cli task +create \
  --summary "Release v2.0" \
  --due "2026-08-15" \
  --data '{"is_milestone":true}'

# Preview the API call without executing
lark-cli task +create --summary "Test Task" --dry-run
```

## Parameters

| Parameter | Required | Description |
|-----------|----------|-------------|
| `--summary <text>` | Yes | The title or summary of the task |
| `--description <text>` | No | Detailed description of the task |
| `--assignee <id>` | No | Assignee ID. Use user `open_id` like `ou_xxx` for people, or app ID like `cli_xxx` for apps. |
| `--follower <id>` | No | Follower ID. Use user `open_id` like `ou_xxx` for people, or app ID like `cli_xxx` for apps. |
| `--due <time>` | No | Due date. Supports ISO 8601, `YYYY-MM-DD`, relative time (e.g., `+2d`), or ms timestamp. `YYYY-MM-DD` and relative time will automatically set it as an all-day task. |
| `--tasklist-id <id>` | No | The GUID of the tasklist, or a full AppLink URL (the CLI will automatically extract the `guid` parameter from the URL). |
| `--idempotency-key <key>` | No | Client token to ensure idempotency of the request. |
| `--data <json>` | No | JSON object merged into the task create request for API fields without dedicated flags, such as `{"is_milestone":true}`. Explicit named flags override same-named fields in this object. |
| `--dry-run` | No | Preview the API call (JSON payload) without actually creating the task. |

> **Required:** If `task +create` has no dedicated flag for a field requested by the user, first inspect `lark-cli schema task.tasks.create`, then add that field to `--data` using the exact field name, type, and nesting from the Meta API request-body schema. Do not omit requested fields or guess their JSON shape. Keep fields already supplied through dedicated flags out of `--data`.

Prefer this shortcut over the raw `tasks create` command when `--data` can express the request. Do not assume that other shortcuts support `--data`; check each shortcut's `--help` output first.

## Workflow

1. Confirm with the user: task summary, due date, assignee, and tasklist if necessary.
   - **Crucial Rule for Assignee**: If the user explicitly or implicitly says "create a task for me" (给我创建一个任务), or "help me create a task" (帮我新建/创建一个任务), you MUST assign the task to the current logged-in user. You can get the current user's `open_id` by executing `lark-cli auth status` (it already outputs JSON by default, so do not add `--json`) or `lark-cli contact +get-user` first, extracting `.identities.user.openId` (from `auth status`) or `.data.user.open_id` (from `contact +get-user`), and then passing it to the `--assignee` parameter.
2. Execute `lark-cli task +create --summary "..." ...`
3. Judge success by `ok == true` in the stdout JSON (the success envelope has no `code` field — do not test `code == 0`), then report the result: task ID (`data.guid`) and summary.

Example success response:

```json
{
  "ok": true,
  "identity": "user",
  "data": {
    "guid": "e297d3d0-4b60-4a5f-a4d4-xxxxxxxxxxxx",
    "url": "https://applink.larkoffice.com/client/todo/detail?guid=e297d3d0-4b60-4a5f-a4d4-xxxxxxxxxxxx"
  }
}
```

> [!CAUTION]
> This is a **Write Operation** -- You must confirm the user's intent before executing.

## References

- [lark-task](../SKILL.md) -- All task commands
- [lark-shared](../../lark-shared/SKILL.md) -- Authentication and global parameters

# task +followers

> **Prerequisites:** Please read `../lark-shared/SKILL.md` to understand authentication, global parameters, and security rules.

Manage task followers. Add or remove followers from an existing task.

## Recommended Commands

```bash
# Add a follower
lark-cli task +followers --task-id "<task_guid>" --add "ou_aaa"

# Add an app follower
lark-cli task +followers --task-id "<task_guid>" --add "cli_xxx"

# Remove a follower
lark-cli task +followers --task-id "<task_guid>" --remove "ou_aaa"
```

## Parameters

| Parameter | Required | Description |
|-----------|----------|-------------|
| `--task-id <guid>` | Yes | The task GUID to modify. For Feishu task applinks, use the `guid` query parameter, not the `suite_entity_num` / display task ID like `t104121`. |
| `--add <ids>` | No | Comma-separated follower IDs. Use user `open_id`s like `ou_xxx` for people, or app IDs like `cli_xxx` for apps. |
| `--remove <ids>` | No | Comma-separated follower IDs. Use user `open_id`s like `ou_xxx` for people, or app IDs like `cli_xxx` for apps. |

## Workflow

1. Confirm the task and followers to add/remove.
2. Execute the command.
3. Report success.

> [!CAUTION]
> This is a **Write Operation** -- You must confirm the user's intent before executing.

# task +get-my-tasks

If the user query only specifies a task name (e.g., "Complete task Lobster No. 1"), use this command to list and search for the task by its summary.

> **Prerequisites:** Please read `../lark-shared/SKILL.md` to understand authentication, global parameters, and security rules.
> 
> **⚠️ Note:** This API must be called with a user identity. **Do NOT use an app identity, otherwise the call will fail.**
>
> **Output rendering note:**
> 1. If you need to present user fields (assignee, creator, etc.), do not only output the raw `id` (e.g. open_id). Also try to resolve and display the user's real name (e.g. via the contact skill) for readability.
> 2. When rendering timestamps (e.g. created time, due time), use the local timezone. Format is 2006-01-02 15:04:05

List tasks assigned to the current user, with support for filtering by completion status, creation time, and due date.
By default, the command will automatically paginate up to 20 times. Use `--page-all` to fetch more (up to 40 pages).

> **Pending vs all tasks:** When `--complete` is not provided, the result contains **both completed and incomplete tasks**.
> For standup / daily-summary / pending-todo scenarios, you **must** pass `--complete=false`; otherwise completed tasks will be surfaced as if they were still pending.

## Recommended Commands

```bash
# Search for a specific task by name
lark-cli task +get-my-tasks --query "Lobster No. 1"

# Get all my tasks, both completed and incomplete (fetches up to 20 pages by default)
lark-cli task +get-my-tasks

# Pending-only: my incomplete tasks (use this for standup/daily-summary)
lark-cli task +get-my-tasks --complete=false

# Pending-only with a due-date upper bound (e.g. end of today / this week)
lark-cli task +get-my-tasks --complete=false --due-end "2026-03-27T23:59:59+08:00"

# Fetch all my tasks (up to 40 pages)
lark-cli task +get-my-tasks --page-all

# Fetch up to 10 pages
lark-cli task +get-my-tasks --page-limit 10

# Resume from a known page token
lark-cli task +get-my-tasks --page-token "pt_xxx"
```

## Parameters

| Parameter | Required | Description |
|-----------|----------|-------------|
| `--query <string>` | No | Search for tasks by summary. Returns exact matches if any; otherwise returns partial matches. |
| `--complete=<bool>` | No | Optional. If not provided, it fetches all tasks (both incomplete and completed). Set to `true` to fetch only completed tasks, or `false` for incomplete tasks. |
| `--created_at <string>` | No | Query tasks created after this time. Supports date: `YYYY-MM-DD`, relative: `-2d`, or ms timestamp. |
| `--due-start <string>` | No | Query tasks with a due date after this time. Supports date: `YYYY-MM-DD`, relative: `-2d`, or ms timestamp. |
| `--due-end <string>` | No | Query tasks with a due date before this time. Supports date: `YYYY-MM-DD`, relative: `-2d`, or ms timestamp. |
| `--page-all` | No | Automatically paginate through all pages (max 40). |
| `--page-limit <int>` | No | Max page limit (default 20). |
| `--page-token <string>` | No | Start from the specified page token (useful for resuming a previous query). |

## Workflow

1. Determine the filters based on the user's request.
2. Execute the command. The CLI will automatically loop up to the specified limit (default 20, or 40 with `--page-all`) to fetch records.
3. Show the results (ID, summary, due time, and created date).

# task +get-related-tasks

> **Prerequisites:** Please read `../lark-shared/SKILL.md` to understand authentication, global parameters, and security rules.
>
> **⚠️ Note:** This API must be called with a user identity. **Do NOT use an app identity, otherwise the call will fail.**
>
> **Pagination / Time Cursor Rule:**
> In `+get-related-tasks`, `page_token` is the task `updated_at` cursor in microseconds.
>
> **Execution Priority:**
> 1. If the request contains a start/end time boundary (for example, "今年以来", "最近一个月", "从 3 月 1 日开始"), first convert the **start time** boundary to a microsecond `page_token` and query from that token.
> 2. Continue pagination using returned `page_token` until `has_more=false`, but never exceed 40 total page fetches.
> 3. Do NOT default to `--page-all` for time-bounded queries.
>
> Only use `--page-all` from the beginning when:
> 1. the user explicitly asks for a full scan of all related tasks, or
> 2. no time boundary can be inferred from the request.

List tasks related to the current user.

## Recommended Commands

```bash
# List all related tasks
lark-cli task +get-related-tasks

# List incomplete related tasks starting from a page token
lark-cli task +get-related-tasks --include-complete=false --page-token "1752730590582902"

# Show only tasks created by me
lark-cli task +get-related-tasks --created-by-me
```

## Parameters

| Parameter | Required | Description |
|-----------|----------|-------------|
| `--include-complete=<bool>` | No | Default behavior includes completed tasks. Set to `false` to keep only incomplete tasks. |
| `--page-all` | No | Automatically paginate through all pages (max 40). |
| `--page-limit <int>` | No | Max page limit (default 20). |
| `--page-token <string>` | No | Start from the specified page token. This token is the task's last update time cursor in microseconds. |
| `--created-by-me` | No | Keep only tasks whose creator is the current user. This is a client-side filter applied after fetching related-task pages. |
| `--followed-by-me` | No | Keep only tasks followed by the current user. This is a client-side filter applied after fetching related-task pages. |

> **Page Token Note:** In `+get-related-tasks`, the `page_token` is a microsecond-level cursor representing the task's last update time. For example, `1752730590582902` should be treated as an updated-at cursor, not a task ID.
>
> **Pagination Note for Client-side Filters:** When `--created-by-me` or `--followed-by-me` is used, filtering happens locally after each upstream related-task page is fetched. The returned `has_more` and `page_token` still describe the upstream cursor, so later pages may contain more matching tasks, or may contain none.

## Workflow

1. Determine whether the user needs all related tasks or a filtered subset.
2. Execute `lark-cli task +get-related-tasks ...`
3. Report the matching tasks and, if present, the next `page_token`.

# task +reminder

> **Prerequisites:** Please read `../lark-shared/SKILL.md` to understand authentication, global parameters, and security rules.
> **Priority:** For creating or modifying task reminder times, prioritize using this `+reminder` shortcut over other task update methods. It provides a more reliable and direct way to manage reminders.

Manage task reminders. Set new reminders or remove existing ones. Note that setting a task reminder requires a due date.

## Recommended Commands

```bash
# Set a reminder (e.g., 30 minutes before due)
lark-cli task +reminder --task-id "<task_guid>" --set "30"

# Set a reminder (e.g., 1 hour before due)
lark-cli task +reminder --task-id "<task_guid>" --set "1h"

# Remove all reminders
lark-cli task +reminder --task-id "<task_guid>" --remove "true"
```

## Parameters

| Parameter | Required | Description |
|-----------|----------|-------------|
| `--task-id <guid>` | Yes | The task GUID to modify. For Feishu task applinks, use the `guid` query parameter, not the `suite_entity_num` / display task ID like `t104121`. |
| `--set <val>` | No | Relative fire minutes before the due time. Supports numbers (e.g., `30`) or units (e.g., `15m`, `1h`, `1d`). |
| `--remove <bool>` | No | If set to `true`, removes all existing reminders from the task. |

## Workflow

1. Confirm the task and reminder action.
2. Execute the command.
3. Report success.

> [!CAUTION]
> This is a **Write Operation** -- You must confirm the user's intent before executing.

# task +reopen

> **Prerequisites:** Please read `../lark-shared/SKILL.md` to understand authentication, global parameters, and security rules.

Reopen a previously completed task.

## Recommended Commands

```bash
# Reopen a task
lark-cli task +reopen --task-id "<task_guid>"
```

## Parameters

| Parameter | Required | Description |
|-----------|----------|-------------|
| `--task-id <guid>` | Yes | The task GUID to reopen. For Feishu task applinks, use the `guid` query parameter, not the `suite_entity_num` / display task ID like `t104121`. |

## Workflow

1. Confirm the task to reopen.
2. Execute the command.
3. Report success.

> [!CAUTION]
> This is a **Write Operation** -- You must confirm the user's intent before executing.

# task +search

> **Prerequisites:** Please read `../lark-shared/SKILL.md` to understand authentication, global parameters, and security rules.
>
> **⚠️ Note:** This API must be called with a user identity. **Do NOT use an app identity, otherwise the call will fail.**

Search tasks by keyword and optional filters.

## Recommended Commands

```bash
# Search by keyword
lark-cli task +search --query "test"

# Search incomplete tasks assigned to specific users
lark-cli task +search --assignee "ou_xxx,ou_yyy" --completed=false

# Search by due time range
lark-cli task +search --query "release" --due "-1d,+7d"
```

## Parameters

| Parameter | Required | Description |
|-----------|----------|-------------|
| `--query <string>` | No | Search keyword. If omitted, at least one filter must be provided. |
| `--creator <ids>` | No | Creator open_ids, comma-separated. |
| `--assignee <ids>` | No | Assignee open_ids, comma-separated. |
| `--follower <ids>` | No | Follower open_ids, comma-separated. |
| `--completed=<bool>` | No | Filter by completion state. |
| `--due <range>` | No | Due time range in `start,end` form. Each side supports ISO/date/relative/ms input. |
| `--page-token <string>` | No | Page token for pagination. |
| `--page-all` | No | Automatically paginate through all pages (max 40). |
| `--page-limit <int>` | No | Max page limit (default 20). |

## Workflow

1. Build the keyword and filters from the user's request.
2. Execute `lark-cli task +search ...`
3. Report the matched tasks and include the next `page_token` if more results exist.

# task +set-ancestor

> **Prerequisites:** Please read `../lark-shared/SKILL.md` to understand authentication, global parameters, and security rules.

Set a parent task for a task, or clear the parent to make it independent.

## Recommended Commands

```bash
# Set a parent task
lark-cli task +set-ancestor --task-id "guid_1" --ancestor-id "guid_2"

# Clear the parent task
lark-cli task +set-ancestor --task-id "guid_1"
```

## Parameters

| Parameter | Required | Description |
|-----------|----------|-------------|
| `--task-id <guid>` | Yes | The task GUID to update. |
| `--ancestor-id <guid>` | No | The parent task GUID. Omit it to clear the ancestor. |

## Workflow

1. Confirm the child task and, if applicable, the ancestor task.
2. Execute `lark-cli task +set-ancestor ...`
3. Report the updated task GUID and whether the ancestor was set or cleared.

> [!CAUTION]
> This is a **Write Operation** -- You must confirm the user's intent before executing.

# task +tasklist-create

> **Prerequisites:** Please read `../lark-shared/SKILL.md` to understand authentication, global parameters, and security rules.

Create a new tasklist, and optionally batch create tasks within it.

## Recommended Commands

```bash
# Create an empty tasklist
lark-cli task +tasklist-create --name "Q1 Goals"

# Create a tasklist and add members
lark-cli task +tasklist-create --name "Project A" --member "ou_xxx,ou_yyy"

# Create a tasklist and batch create tasks within it
lark-cli task +tasklist-create --name "Launch Checklist" --data '[{"summary": "Code Review", "assignee": "ou_aaa"}, {"summary": "Deploy", "assignee": "ou_bbb"}]'
```

## Parameters

| Parameter | Required | Description |
|-----------|----------|-------------|
| `--name <text>` | Yes | The name of the tasklist. |
| `--member <ids>` | No | Comma-separated list of user `open_id`s to add as editors. |
| `--data <json>` | No | JSON array of task definitions to create and add to the tasklist automatically. |

## Workflow

1. Confirm the tasklist name, members, and tasks (if any).
2. Execute the command `lark-cli task +tasklist-create ...`.
3. Report success, including the new tasklist ID and the result of the batch task creation.

> [!CAUTION]
> This is a **Write Operation** -- You must confirm the user's intent before executing.

# task +tasklist-members

> **Prerequisites:** Please read `../lark-shared/SKILL.md` to understand authentication, global parameters, and security rules.

Manage tasklist members (editors/owners).

## Recommended Commands

```bash
# Add a member
lark-cli task +tasklist-members --tasklist-id "tl_xxx" --add "ou_aaa"

# Remove a member
lark-cli task +tasklist-members --tasklist-id "tl_xxx" --remove "ou_aaa"

# Replace all members exactly
lark-cli task +tasklist-members --tasklist-id "tl_xxx" --set "ou_aaa,ou_bbb"
```

## Parameters

| Parameter | Required | Description |
|-----------|----------|-------------|
| `--tasklist-id <id>` | Yes | The GUID of the tasklist, or a full AppLink URL. |
| `--add <ids>` | No | Comma-separated list of user `open_id`s to add as members. |
| `--remove <ids>` | No | Comma-separated list of user `open_id`s to remove from members. |
| `--set <ids>` | No | Comma-separated list of user `open_id`s to exactly set as members (replaces all existing). |

## Workflow

1. Confirm the tasklist and members to add/remove/set.
2. Execute the command.
3. Report success.

> [!CAUTION]
> This is a **Write Operation** -- You must confirm the user's intent before executing.

# task +tasklist-search

> **Prerequisites:** Please read `../lark-shared/SKILL.md` to understand authentication, global parameters, and security rules.
>
> **⚠️ Note:** This shortcut uses tasklist search followed by tasklist detail queries to render the final output.

Search tasklists by keyword and optional filters.

## Recommended Commands

```bash
# Search by keyword
lark-cli task +tasklist-search --query "测试"

# Search tasklists created by specific users
lark-cli task +tasklist-search --creator "ou_xxx,ou_yyy"

# Search by creation time range
lark-cli task +tasklist-search --query "Q2" --create-time "-30d,+0d"
```

## Parameters

| Parameter | Required | Description |
|-----------|----------|-------------|
| `--query <string>` | No | Search keyword. If omitted, at least one filter must be provided. |
| `--creator <ids>` | No | Creator open_ids, comma-separated. |
| `--create-time <range>` | No | Creation time range in `start,end` form. Each side supports ISO/date/relative/ms input. |
| `--page-token <string>` | No | Page token for pagination. |
| `--page-all` | No | Automatically paginate through all pages (max 40). |
| `--page-limit <int>` | No | Max page limit (default 20). |

## Workflow

1. Build the search keyword and filters from the user's request.
2. Execute `lark-cli task +tasklist-search ...`
3. Report the matched tasklists and the next `page_token` if more results exist.

# task +tasklist-task-add

> **Prerequisites:** Please read `../lark-shared/SKILL.md` to understand authentication, global parameters, and security rules.

Add existing tasks to a tasklist.

## Recommended Commands

```bash
# Add a single task to a tasklist
lark-cli task +tasklist-task-add --tasklist-id "<tasklist_guid>" --task-id "<task_guid>"

# Add multiple tasks to a tasklist
lark-cli task +tasklist-task-add --tasklist-id "<tasklist_guid>" --task-id "<task_guid>,<another_task_guid>,<third_task_guid>"

# Add a task to a specific section in the tasklist
lark-cli task +tasklist-task-add \
  --tasklist-id "<tasklist_guid>" \
  --task-id "<task_guid>" \
  --section-guid "<section_guid>"
```

## Parameters

| Parameter | Required | Description |
|-----------|----------|-------------|
| `--tasklist-id <guid>` | Yes | The GUID of the tasklist, or a full AppLink URL. |
| `--task-id <guids>` | Yes | Comma-separated list of task GUIDs to add to the tasklist. For Feishu task applinks, use each task's `guid` query parameter, not the `suite_entity_num` / display task ID like `t104121`. |
| `--section-guid <guid>` | No | The GUID of the custom section to add the tasks to. If omitted, tasks will be added to the default section. |

## Workflow

1. Confirm the tasklist and the tasks to add.
2. Execute the command `lark-cli task +tasklist-task-add ...`.
3. Report the result (successful vs failed tasks).

> [!CAUTION]
> This is a **Write Operation** -- You must confirm the user's intent before executing.

# task +update

> **Prerequisites:** Please read `../lark-shared/SKILL.md` to understand authentication, global parameters, and security rules.

Update an existing task in Lark.

## Recommended Commands

```bash
# Update task summary
lark-cli task +update --task-id "<task_guid>" --summary "New Summary"

# Update multiple tasks' due dates
lark-cli task +update --task-id "<task_guid>,<another_task_guid>" --due "+2d"

# A task applink is accepted directly; the CLI extracts its guid query value
lark-cli task +update --task-id "https://applink.larksuite.com/client/todo/task?guid=<task_guid>" --summary "New Summary"

# Update with JSON data
lark-cli task +update --task-id "<task_guid>" --data '{"description": "New description"}'
```

## Parameters

| Parameter | Required | Description |
|-----------|----------|-------------|
| `--task-id <guid-or-applink>` | Yes | Task OpenAPI GUID or a task applink containing `guid=`. Comma-separated GUIDs/applinks are supported for multiple tasks. Display task IDs such as `t104121` / `suite_entity_num` are rejected. |
| `--summary <text>` | No | New summary/title for the task. |
| `--description <text>` | No | New description for the task. |
| `--due <time>` | No | New due date (supports relative time). |
| `--data <json>` | No | JSON payload for fields to update. |

## Workflow

1. Confirm with the user the tasks to update and the fields.
2. Execute `lark-cli task +update --task-id "..." ...`
3. Read `data.updated_fields` and `data.tasks[].confirmed` from the result and report only the fields confirmed by the server.
4. Do not routinely call `task tasks get` after the update when `confirmed` already contains the required state. Query details only if a required field is absent or the user explicitly asks for a full verification.

> [!CAUTION]
> This is a **Write Operation** -- You must confirm the user's intent before executing.

# task +upload-attachment

> **Prerequisites:** Please read `../lark-shared/SKILL.md` to understand authentication, global parameters, and security rules.

Upload a single local file as an attachment to a task (or any resource type accepted by the Task attachment endpoint). Max file size per upload is **50 MB**. For task agents, use `--resource-type=task_delivery`.

## Recommended Commands

```bash
# Upload a local file as a task attachment (relative path required)
lark-cli task +upload-attachment \
  --resource-id "<task_guid>" \
  --file "./report.pdf"

# Pass a Feishu task applink instead of a raw guid — the guid is extracted automatically
lark-cli task +upload-attachment \
  --resource-id "https://applink.feishu.cn/client/todo/task?guid=<task_guid>" \
  --file "./note.md"

# Explicit resource type / user id type
lark-cli task +upload-attachment \
  --resource-id "<task_guid>" \
  --resource-type task \
  --user-id-type open_id \
  --file "./design.png"

# Upload a local file to a task agent
lark-cli task +upload-attachment \
  --resource-id "4b113c53-a68b-419f-8bd0-c9c532a3285a" \
  --file "./飞书.zip" \
  --resource-type task_delivery
```

## Parameters

| Parameter | Required | Description |
|-----------|----------|-------------|
| `--resource-id <guid_or_applink>` | Yes | Target resource GUID. Accepts a raw task GUID or a Feishu task applink URL (`.../client/todo/task?guid=...`); the `guid` query parameter is extracted automatically. Do not use `suite_entity_num` / display IDs like `t104121`. |
| `--file <path>` | Yes | Local file path to upload. Must be a relative path within the current working directory; absolute paths and paths escaping the cwd are rejected. Single file only, ≤ 50 MB. |
| `--resource-type <type>` | No | Owning resource type. Defaults to `task`. Use `task_delivery` when uploading to task agents. |
| `--user-id-type <type>` | No | User ID type for the request. Defaults to `open_id`. |

## Workflow

1. Confirm the target task GUID (or applink) and the local file path with the user.
2. Ensure the file is within the current working directory and its size is ≤ 50 MB; otherwise ask the user to move/split the file.
3. Determine if this is a task agent: if yes, add `--resource-type task_delivery`.
4. Execute `lark-cli task +upload-attachment --resource-id "..." --file "..."`.
5. Report the returned attachment record. The output exposes all fields returned by the API (e.g. `guid`, `name`, `size`, `url`, `uploader`, ...); always surface the attachment `guid` and, if present, the `url` so the user can jump to the attachment directly.

## Output

The command returns the single created attachment record as a flat JSON object — every field returned by the API (`guid`, `name`, `size`, `url`, `resource_type`, `resource_id`, `uploader`, ...) is preserved verbatim. Pretty mode also prints a human-readable summary with the resource, file name, size, and attachment GUID.

> [!CAUTION]
> This is a **Write Operation** -- You must confirm the user's intent before executing.

> [!NOTE]
> The Task attachment upload endpoint accepts exactly one file per call. To upload multiple files, invoke the shortcut once per file.

---
name: lark-task
version: 1.0.0
description: "飞书任务：管理任务、清单和任务智能体。创建待办任务、查看和更新任务状态、拆分子任务、组织任务清单、分配协作成员、上传任务附件、注册或注销任务智能体、更新任务智能体的主页数据、写入智能体任务记录。当用户需要创建待办事项、查看任务列表、跟踪任务进度、管理项目清单或给他人分配任务、为任务上传附件文件、注册注销任务智能体、更新智能体主页数据、写入任务记录时使用。"
metadata:
  requires:
    bins: ["lark-cli"]
  cliHelp: "lark-cli task --help"
---

# task (v2)

**CRITICAL — 开始前 MUST 先用 Read 工具读取 [`../lark-shared/SKILL.md`](../lark-shared/SKILL.md)，其中包含认证、权限处理**

## 命令选择与渐进式发现（必读）

执行任何 Task 命令前，必须先确认能力真实存在，禁止根据用户意图自行拼接或猜测 `+<verb>`：

1. 先将用户意图与下方 Shortcut 表精确匹配。只有表中明确列出的 shortcut 才可直接选择；参数不确定时读取对应 reference 或运行该 shortcut 的 `--help`。
2. 没有精确匹配、或无法确认当前版本是否支持时，先运行 `lark-cli task --help`，以当前 CLI 输出的命令列表为准。
3. help 中存在匹配 shortcut 时，使用 help 列出的完整 shortcut token（例如 `+create`）运行 `lark-cli task <shortcut> --help`，再按真实 flag 执行。
4. help 中没有匹配 shortcut 时，不得尝试相似的 `+<verb>`；从 help 中选择原生 resource，运行 `lark-cli task <resource> --help` 确认 method，再运行 `lark-cli schema task.<resource>.<method>` 获取参数结构，最后调用 `lark-cli task <resource> <method> ...`。
5. 遇到 `unknown_subcommand` 时必须停止猜测或尝试变体，回到第 2 步重新发现能力。

shortcut 名称只能来自本 Skill 的 Shortcut 表或 `lark-cli task --help`；原生 resource/method 以逐级 help 为准，参数名、类型和嵌套结构以 method schema 为准。

> **任务搜索技巧**：先区分用户是否**特地指定使用搜索 skill**，以及是否真的提供了**查询关键字**（例如任务名称、关键词、片段描述）。如果用户特地指定使用搜索 skill，或明确给出了任务查询关键字，则目标是**任务**时优先使用 `+search`。如果用户没有特地指定使用搜索 skill，且意图里没有查询关键字，只有范围条件（例如“今年以来”“已完成”“由我创建”“我关注的”），并且使用 `+search` 与 `+get-related-tasks` / `+get-my-tasks` 都能达到目的时，应优先使用列表型能力，而不是搜索型能力。其中，“与我相关 / 我关注的 / 由我创建”等优先考虑 `+get-related-tasks`；“我负责的 / 分配给我”的列表优先考虑 `+get-my-tasks`。不要把时间范围词（例如“今年以来”）本身误当成 `query` 去走搜索。
> **任务搜索相关性提示**：`+search` 当前不会自动判断搜索结果与搜索发起人的相关性。如果用户明确要求搜索“与我相关”的任务，必须先识别具体关系，获取当前用户的 `open_id`，并显式传入对应的 `--assignee`（负责人）、`--creator`（创建人）或 `--follower`（关注人）过滤条件；不能只依赖 `query` 期待自动返回与当前用户相关的任务。
> **任务清单搜索技巧**：任务清单也遵循同样的判断逻辑。先区分用户是否**特地指定使用搜索 skill**，以及是否真的提供了**清单查询关键字**（例如清单名称、关键词、片段描述）。如果用户特地指定使用搜索 skill，或明确给出了清单查询关键字，则优先使用 `+tasklist-search`。如果用户没有特地指定使用搜索 skill，且意图里没有查询关键字，只有范围条件（例如“由我创建的任务清单”“今年以来创建的清单”），并且使用搜索或原生列取清单都能达到目的时，应优先使用原生 `tasklists.list` 接口列取清单（先 `schema task.tasklists.list`，再 `lark-cli task tasklists list --as user ...`），再按 `creator`、`created_at` 等字段做本地筛选和分页控制。
> **意图区分补充**：像“搜索飞书中今年以来我关注的任务”这类表达，虽然字面带有“搜索”，但如果没有真正的查询关键字，且本质是在限定“与我相关 + 时间范围”，则应优先走 `+get-related-tasks`；像“搜索飞书中由我创建的任务清单”这类表达，如果没有清单关键字，且本质是在限定“清单范围 + 创建者”，则应优先走原生 `tasklists.list` 后筛选，而不是直接走搜索型 shortcut。
> **用户身份识别**：在用户身份（user identity）场景下，如果用户提到了“我”（例如“分配给我”、“由我创建”），请默认获取当前登录用户的 `open_id` 作为对应的参数值。
> **术语理解 — 待办 disambiguation（必读）**：
> - 用户提到「待办 / todo / 任务」时，**先判断归属**，不要默认走本 skill。
> - **走 [lark-meeting](../lark-meeting/SKILL.md) 的 `minutes +todo`**（禁止本 skill）：上下文含 **妙记 / 会议纪要 / minute_token / 妙记 URL**（`/minutes/`）；或「在某某妙记里新建/修改待办」「妙记 AI 待办」「会议录制里的待办」。
> - **走本 skill（lark-task）**：任务清单、分配给我、项目待办、截止日期/提醒、子任务、任务清单成员；或 applink 含 `client/todo/task?guid=`；或明确说「飞书任务」「任务中心」「我的任务清单」。
> - **禁止**：用户要在妙记里加待办时，**不要**调用 `task tasklists list`、`task +create` 或任何 task 命令去「找清单再放任务」。
> **友好输出**：在输出任务（或清单）的执行结果给用户时，建议同时提取并输出命令返回结果中的 `url` 字段（任务链接），以便用户可以直接点击跳转查看详情。

> **创建/更新注意**：
> 1. 只有在设置了 `due`（截止时间）的情况下，才能设置 `repeat_rule`（重复规则）和 `reminder`（提醒时间）。
> 2. 若同时设置了 `start`（开始时间）和 `due`（截止时间），开始时间必须小于或等于截止时间。
> 3. 使用 tenant_access_token（应用身份）时，无法跨租户添加任务成员。

> **查询注意**：
> 1. 在输出任务详情时，如果需要渲染负责人、创建人等人员字段，除了展示 `id` (例如 open_id) 外，还必须通过其他方式（例如调用通讯录技能）尝试获取并展示这个人的真实名字，以便用户更容易识别。
> 2. 在输出清单详情时，如果需要渲染 owner、member、角色成员等人员字段，也必须像任务成员展示一样，除了展示 `id` 外，尽量解析并展示对应人员的真实名字。
> 3. 在输出任务或清单详情时，如果需要渲染创建时间、截止时间等字段，需要使用本地时区来渲染（格式为2006-01-02 15:04:05）。

> **Task GUID 定义**：
> Task OpenAPI 中用于更新/操作任务的 `guid` 是任务的全局唯一标识（GUID），不是客户端展示的任务编号（例如 `t104121` / `suite_entity_num`）。
> 对于 Feishu 的任务 applink（例如 `.../client/todo/task?guid=...`），必须使用 URL query 里的 `guid` 参数作为 task guid。

> **从任务清单定位并修改任务的最短路径**：
> 1. 已知任务清单 GUID 时直接使用，不要先搜索；已知任务清单 applink 时，取 URL query 中的 `guid` 作为 `tasklist_guid`。
> 2. 只有清单名称或关键词、没有 GUID/applink 时，才调用一次 `+tasklist-search` 解析目标清单。
> 3. 按原生 API 规则先执行 `lark-cli schema task.tasklists.tasks`，再执行 `lark-cli task tasklists tasks --params '{"tasklist_guid":"<tasklist_guid>"}' --as user`。
> 4. 从清单任务结果中取任务的 `guid`，直接传给 `+update` 或 `+complete`；禁止传客户端展示编号（例如 `t104121`）。这两个 shortcut 也可直接接收包含 `guid=` 的任务 applink。
> 5. `+update` 返回 `updated_fields` 和每个任务的服务端 `confirmed` 字段；`+complete` 返回 `status`、`completed_at`、`already_completed`。这些字段已确认目标状态时，不要例行追加 `tasks get`；仅在服务端未返回所需字段或用户明确要求完整复核时再查询详情。

| Shortcut | 说明 |
|----------|------|
| [`+create`](references/lark-task-create.md) | create a task |
| [`+update`](references/lark-task-update.md) | update task attributes |
| [`+set-ancestor`](references/lark-task-set-ancestor.md) | set or clear a task ancestor |
| [`+comment`](references/lark-task-comment.md) | add a comment to a task |
| [`+complete`](references/lark-task-complete.md) | mark a task as complete |
| [`+reopen`](references/lark-task-reopen.md) | reopen a completed task |
| [`+assign`](references/lark-task-assign.md) | assign or remove task members |
| [`+followers`](references/lark-task-followers.md) | manage task followers |
| [`+reminder`](references/lark-task-reminder.md) | manage task reminders |
| [`+get-my-tasks`](references/lark-task-get-my-tasks.md) | List tasks assigned to me |
| [`+get-related-tasks`](references/lark-task-get-related-tasks.md) | list tasks related to me |
| [`+search`](references/lark-task-search.md) | search tasks |
| [`+upload-attachment`](references/lark-task-upload-attachment.md) | upload a local file as an attachment to a task |
| [`+tasklist-create`](references/lark-task-tasklist-create.md) | create a tasklist and optionally add tasks |
| [`+tasklist-search`](references/lark-task-tasklist-search.md) | search tasklists |
| [`+tasklist-task-add`](references/lark-task-tasklist-task-add.md) | add tasks to a tasklist |
| [`+tasklist-members`](references/lark-task-tasklist-members.md) | manage tasklist members |

## API Resources

```bash
lark-cli schema task.<resource>.<method>   # 调用 API 前必须先查看参数结构
lark-cli task <resource> <method> [flags] # 调用 API
```

> **重要**：使用原生 API 时，必须先运行 `schema` 查看 `--data` / `--params` 参数结构，不要猜测字段格式。

### tasks

  - `create` — 创建任务
  - `delete` — 删除任务
  - `get` — 获取任务详情
  - `list` — 列取任务列表
  - `patch` — 更新任务

### tasklists

  - `add_members` — 添加清单成员
  - `create` — 创建清单
  - `delete` — 删除清单
  - `get` — 获取清单详情
  - `list` — 获取清单列表
  - `patch` — 更新清单
  - `remove_members` — 移除清单成员
  - `tasks` — 获取清单任务列表

### subtasks

  - `create` — 创建子任务
  - `list` — 获取任务的子任务列表

### members

  - `add` — 添加任务成员
  - `remove` — 移除任务成员

### sections

  - `create` — 创建自定义分组
  - `delete` — 删除自定义分组
  - `get` — 获取自定义分组详情
  - `list` — 获取自定义分组列表
  - `patch` — 更新自定义分组
  - `tasks` — 获取自定义分组任务列表

### custom_fields

  - `create` — 创建自定义字段
  - `get` — 获取自定义字段详情
  - `patch` — 更新自定义字段
  - `list` — 获取自定义字段列表
  - `add` — 将自定义字段加入资源
  - `remove` — 将自定义字段移出资源

### custom_field_options

  - `create` — 创建自定义字段选项
  - `patch` — 更新自定义字段选项

### agent

  - `update_agent_profile` — 更新任务代理的主页内容数据。
  - `register_agent` — 注册AI 智能体

### agent_task_step_info

  - `append_task_steps` — 写入任务记录。

## 权限表

| 方法 | 所需 scope |
|------|-----------|
| `tasks.create` | `task:task:write` |
| `tasks.delete` | `task:task:write` |
| `tasks.get` | `task:task:read` |
| `tasks.list` | `task:task:read` |
| `tasks.patch` | `task:task:write` |
| `tasklists.add_members` | `task:tasklist:write` |
| `tasklists.create` | `task:tasklist:write` |
| `tasklists.delete` | `task:tasklist:write` |
| `tasklists.get` | `task:tasklist:read` |
| `tasklists.list` | `task:tasklist:read` |
| `tasklists.patch` | `task:tasklist:write` |
| `tasklists.remove_members` | `task:tasklist:write` |
| `tasklists.tasks` | `task:tasklist:read` |
| `subtasks.create` | `task:task:write` |
| `subtasks.list` | `task:task:read` |
| `members.add` | `task:task:write` |
| `members.remove` | `task:task:write` |
| `sections.create` | `task:section:write` |
| `sections.delete` | `task:section:write` |
| `sections.get` | `task:section:read` |
| `sections.list` | `task:section:read` |
| `sections.patch` | `task:section:write` |
| `sections.tasks` | `task:section:read` |
| `custom_fields.create` | `task:custom_field:write` |
| `custom_fields.get` | `task:custom_field:read` |
| `custom_fields.patch` | `task:custom_field:write` |
| `custom_fields.list` | `task:custom_field:read` |
| `custom_fields.add` | `task:custom_field:write` |
| `custom_fields.remove` | `task:custom_field:write` |
| `custom_field_options.create` | `task:custom_field:write` |
| `custom_field_options.patch` | `task:custom_field:write` |
| `agent.update_agent_profile` | `task:task:write` |
| `agent.register_agent` | `task:task:write` |
| `agent_task_step_info.append_task_steps` | `task:task:write` |