task (v2)

CRITICAL, Before starting, you MUST use the Read tool to read ../lark-shared/SKILL.md, which contains information on authentication and permission handling.

Task search tips: First, distinguish whether the user has specifically requested to use the search skill and whether they have actually provided query keywords (e.g., task name, keywords, snippet description). If the user specifically requests the search skill or provides clear task query keywords, prioritize +search when the target is a task. If the user does not specifically request the search skill and there are no query keywords in the intent, but only scope conditions (e.g., "since this year", "completed", "created by me", "followed by me"), and both +search and +get-related-tasks / +get-my-tasks can achieve the goal, prioritize list-based capabilities instead of search-based ones. Specifically, for "related to me / followed by me / created by me", prioritize +get-related-tasks; for lists of "tasks I am responsible for / assigned to me", prioritize +get-my-tasks. Do not mistake time range terms (e.g., "since this year") as the query for a search. Tasklist search tips: Tasklists follow the same logic. First, distinguish whether the user has specifically requested to use the search skill and whether they have actually provided list query keywords (e.g., list name, keywords, snippet description). If the user specifically requests the search skill or provides clear list query keywords, prioritize +tasklist-search. If the user does not specifically request the search skill and there are no query keywords in the intent, but only scope conditions (e.g., "tasklists created by me", "lists created since this year"), and both search or native list retrieval can achieve the goal, prioritize the native tasklists.list interface (first schema task.tasklists.list, then lark-cli task tasklists list --as user ...), and then perform local filtering and pagination control based on fields like creator or created_at. Intent distinction supplement: Expressions like "search for tasks followed by me since this year in Feishu", although containing the word "search", should prioritize +get-related-tasks if there are no real query keywords and the essence is limiting "related to me + time range". Expressions like "search for tasklists created by me in Feishu", if there are no list keywords and the essence is limiting "list scope + creator", should prioritize native tasklists.list followed by filtering, rather than going directly to a search shortcut. User identity identification: In user identity scenarios, if the user mentions "me" (e.g., "assigned to me", "created by me"), please default to obtaining the open_id of the currently logged-in user as the corresponding parameter value. Terminology understanding: If the user mentions "todo", consider whether they mean "task" and prioritize using the commands provided by this Skill to handle it. Friendly output: When outputting the execution results of tasks (or lists) to the user, it is recommended to extract and output the url field (task link) from the command return results so that the user can click to jump and view details.

Creation/Update notes:

1. repeat_rule (repeat rule) and reminder (reminder time) can only be set if due (due time) is set.
2. If both start (start time) and due (due time) are set, the start time must be less than or equal to the due time.
3. When using tenant_access_token (app identity), you cannot add task members across tenants.

Query notes:

1. When outputting task details, if you need to render fields like assignee or creator, in addition to displaying the id (e.g., open_id), you must try to obtain and display the person's real name through other means (e.g., calling the address book skill) to make it easier for the user to identify.
2. When outputting list details, if you need to render fields like owner, member, or role members, you must also try to resolve and display the corresponding person's real name in addition to the id, just like displaying task members.
3. When outputting task or list details, if you need to render fields like creation time or due time, you need to use the local time zone for rendering (format: 2006-01-02 15:04:05).

Task GUID definition: The guid used in the Task OpenAPI for updating/operating tasks is the globally unique identifier (GUID) of the task, not the task number displayed on the client (e.g., t104121 / suite_entity_num). For Feishu task applinks (e.g., .../client/todo/task?guid=...), you must use the guid parameter in the URL query as the task guid.

API Resources

lark-cli schema task.<resource>.<method>   # Must view parameter structure before calling API
lark-cli task <resource> <method> [flags] # Call API

Important: When using native APIs, you must first run schema to view the --data / --params parameter structure. Do not guess the field format.

tasks

• create, Create task
• delete, Delete task
• get, Get task details
• list, List tasks
• patch, Update task

tasklists

• add_members, Add list members
• create, Create list
• delete, Delete list
• get, Get list details
• list, Get list of lists
• patch, Update list
• remove_members, Remove list members
• tasks, Get list of tasks in a list

subtasks

• create, Create subtask
• list, Get list of subtasks for a task

members

• add, Add task member
• remove, Remove task member

sections

• create, Create custom section
• delete, Delete custom section
• get, Get custom section details
• list, Get list of custom sections
• patch, Update custom section
• tasks, Get list of tasks in a custom section

custom_fields

• create, Create custom field
• get, Get custom field details
• patch, Update custom field
• list, Get list of custom fields
• add, Add custom field to resource
• remove, Remove custom field from resource

custom_field_options

• create, Create custom field option
• patch, Update custom field option

agent

• update_agent_profile, Update task agent profile data.
• register_agent, Register AI agent

agent_task_step_info

• append_task_steps, Write task record.

Permission Table