lark-skill-maker

Creates reusable lark-cli Skills that wrap Feishu OpenAPI operations into SKILL.md files.

1.3k|142|Updated Jun 1, 2026
One-click install
npx skills add https://github.com/MiniMax-AI/minimax-code --skill lark-skill-maker-minimax-ai
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: lark-skill-maker
Source: https://github.com/MiniMax-AI/minimax-code/tree/main/packages/local-runtime/assets/skills/lark-tools/cli-skills/lark-skill-maker
Command: npx skills add https://github.com/MiniMax-AI/minimax-code --skill lark-skill-maker-minimax-ai

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires lark-cli.

What problem does it solve? Turning Feishu (Lark) API operations into repeatable, discoverable Skills requires knowing which lark-cli commands, parameters, and permission scopes to use. This Skill guides the creation of new SKILL.md files that package single API calls or multi-step workflows into reusable units. ## Core Features & Use Cases - API Research Workflow: Inspect registered APIs and Shortcuts with lark-cli <service> --help, query parameter definitions via lark-cli schema, and call any Feishu OpenAPI endpoint through lark-cli api. - SKILL.md Template: Provides a ready-made frontmatter and body template covering commands, multi-step orchestration with data passing, and a permission scope table. - Auth & Safety Guidance: Points to the canonical lark-shared authentication flow for missing scopes and recommends --dry-run previews before write operations. - Use Case: When you repeatedly create Feishu calendar events and then send notifications, use this Skill to package that two-step flow into a new lark-<name> Skill with documented scopes and step-by-step data passing. ## Quick Start Ask the agent to create a new lark-cli Skill that wraps a specific Feishu API operation, such as searching meeting rooms and booking one.

Frequently Asked Questions about lark-skill-maker

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I create a custom Skill for lark-cli?

Create a SKILL.md file under skills/lark-<name>/ with frontmatter containing name, version, description, and a requires bins entry for lark-cli. The body documents the lark-cli commands, multi-step orchestration, and required permission scopes.

How do I find which Feishu APIs lark-cli supports?

Run lark-cli <service> --help to list registered APIs and Shortcuts, then use lark-cli schema <service.resource.method> to inspect parameter definitions. For unregistered endpoints, call them directly with lark-cli api <METHOD> <path>.

What is the difference between lark-cli Shortcuts and the api command?

Shortcuts are high-level encapsulated operations, registered APIs are standard resource methods, and lark-cli api makes raw calls to any Feishu OpenAPI path. The recommended priority is Shortcut first, then registered API, then raw api calls.

How do I handle missing Feishu permission scopes in a Skill?

Document the required scopes in a permission table and point to the canonical authentication template in lark-shared/SKILL.md. The agent runs --recommend for basic scopes or --scope <name> for sensitive ones, using --no-wait --json with background device-code flow.

Can a lark-cli Skill orchestrate multiple API calls?

Yes. The SKILL.md template supports multi-step orchestration where you document how data passes between steps, such as recording an ID from step one and using it in step two, along with rollback and parallelization notes.