How to use the Feishu Attendance Check-in Lookup agent skill

What is an agent skill, and what does this one do?

An agent skill is a folder containing a SKILL.md file. The file gives an AI agent instructions for a specific task, and the agent reads it when a request matches the skill's description. This skill is named lark-attendance and its description is 飞书考勤打卡：查询自己的考勤打卡记录, which means it queries the signed-in user's own attendance check-in records in Feishu.

The skill teaches the agent to call the lark-cli tool with the attendance resource. It covers one operation, user_tasks.query, which queries the check-in records of the signed-in user. The required scope for that operation is attendance:task:readonly, so the skill is read-only and is not meant for viewing or managing other people's attendance.

• Works only on your own records, as the signed-in user.
• Covers one operation: user_tasks.query.
• Requires the attendance:task:readonly authorization scope.

What you need before installing

The skill cannot run without two things: a Feishu or Lark account with authorization for the attendance:task:readonly scope, and the lark-cli command-line tool, which the skill declares as a required binary. The skill also tells the agent to read a shared authentication skill at ../lark-shared/SKILL.md before doing anything else, because that file covers authentication and permission handling. Keep that shared skill available alongside this one.

• A Feishu/Lark account with the attendance:task:readonly scope.
• The lark-cli command-line tool installed and available to the agent.
• The shared authentication skill at ../lark-shared/SKILL.md.

Install the skill from the ZIP

Download the ZIP for this skill from this page and unzip it. The package contains SKILL.md. Then place the folder in one of two locations: your personal skills folder at ~/.claude/skills/, or a project-specific .claude/skills/ folder inside the project you work in. Use the personal folder if you want the skill available everywhere, and the project folder if it should only apply to one codebase.

If you use the Claude apps instead of the file system, you can upload the skill under Settings → Capabilities → Skills.

• Personal install: ~/.claude/skills/
• Project install: .claude/skills/ inside your project
• Claude apps: Settings → Capabilities → Skills

Trigger the skill with a natural request

You do not need to type a command. Ask the agent in plain language about your own attendance check-in records. The agent matches the request to the skill's description and follows its instructions. Because the skill is about your own records, phrase requests in terms of yourself, such as checking your clock-in history.

Before calling any API, the agent is told to run the schema command to look up the parameter structure, rather than guessing field formats. You can also check the tool's own help with lark-cli attendance --help, which the skill metadata references.

• Ask about your own clock-in and clock-out history.
• Ask what check-in records you have for a given period, if the agent supports the filters you need.
• Expect the agent to look up parameter structure first rather than guessing.

Understand the fields the agent fills in for you

The skill tells the agent to fill in two request fields automatically and not ask you for them. The employee_type field is always set to "employee_no", and the user_ids field is always an empty array []. When building --params or --data for a call, the agent injects these values alongside the details you provide.

These fixed values are part of the skill's design, so you should not be prompted for them. If you are a developer, you can inspect the expected structure yourself with the schema command shown below.

• employee_type is always "employee_no".
• user_ids is always [] (an empty array).
• The agent includes these in addition to your own request details.

Safety notes: scripts and network access

This skill runs scripts and uses the network. It calls lark-cli, which contacts Feishu/Lark services on your behalf, so the records are fetched over the network using your signed-in authorization. The skill only requests read access through the attendance:task:readonly scope and is limited to your own records.

Review the authorization you grant, keep the scope limited to what you need, and check that you are comfortable with an agent running lark-cli commands on your machine before you install the skill.

• Runs scripts: yes, via the lark-cli tool.
• Uses network: yes, to query Feishu/Lark on your behalf.
• Scope is read-only and limited to your own records.