Skill Maker

Skill Maker:

Build Custom Skills for lark-cli API Workflows

The Skill Maker (Skill ID: lark-skill-maker) is a specialized skill within the Happycapy Skills platform, designed for rapid creation of custom skills that encapsulate Feishu (Lark) API operations. It enables technical users to package atomic API calls or orchestrate multi-step workflows into reusable command-line instructions, leveraging the full power of the lark-cli tool.

Tagline:
lark-cli api GET /open-apis/vc/v1/rooms --params '{"pagesize":"50"}'

Category:
Design

Source:
Skill Maker on GitHub

What Is Skill Maker?

Skill Maker is a meta-skill for creating new, task-oriented skills based on the lark-cli command-line interface. At its core, a skill in this context is a structured SKILL.md file which documents and operationalizes a set of CLI commands for a specific business purpose. These commands can be direct API invocations, registered resource operations, or higher-level orchestrations (shortcuts).

By standardizing the way skills are defined and consumed, Skill Maker provides a repeatable pattern for transforming Feishu OpenAPI calls into modular, reusable CLI workflows.

Why Use Skill Maker?

Modern API platforms like Feishu offer extensive capabilities, but interacting with them via raw API calls can be tedious and error-prone, especially for non-trivial workflows. Skill Maker addresses several key pain points:

• Abstraction: Encapsulate low-level API details inside a skill, exposing only the necessary parameters and logic.
• Reusability: Package business logic or frequently used API sequences into shareable, versioned skills.
• Automation: Enable AI and humans to trigger complex operations using simple, declarative skill definitions.
• Documentation: The SKILL.md format enforces clear, consistent documentation of each skill's purpose, parameters, and command structure.
• Extensibility: Skills can use registered APIs, shortcuts, or raw OpenAPI calls, and can orchestrate multi-step flows with data passing between steps.

Skill Maker is ideal for teams seeking to standardize their use of Feishu APIs, automate common patterns, or empower AI agents to perform tasks via CLI instructions.

How to Use Skill Maker

1. Understand the CLI Core

Capabilities

Skill Maker leverages the full range of lark-cli features. Here is a summary:

lark-cli <service> <resource> <method>          # Use a registered API
lark-cli <service> +<verb>                      # Use a shortcut (higher-level abstraction)
lark-cli api <METHOD> <path> [--data/--params]  # Call any Feishu OpenAPI endpoint directly
lark-cli schema <service.resource.method>       # View parameter and schema definitions

Command precedence: Shortcuts are preferred, followed by registered APIs, then raw OpenAPI calls.

2. Research Available

APIs

Before creating a skill, you must identify the correct API endpoints and understand their parameters:

## List available resources and shortcuts
lark-cli <service> --help

## View parameter schema for a resource
lark-cli schema <service.resource.method>

## Call an unregistered API endpoint directly
lark-cli api GET /open-apis/vc/v1/rooms --params '{"page_size":"50"}'
lark-cli api POST /open-apis/vc/v1/rooms/search --data '{"query":"5F"}'

If the required operation is not covered by a registered API or shortcut, use the lark-openapi-explorer skill to explore Feishu’s official documentation, identify the raw OpenAPI endpoint, and determine the required scopes.

3. Write the

SKILL.md File

A new skill is defined by a SKILL.md file placed in the skills/lark-<name>/ directory. The minimum structure is:

Frontmatter:

---
name: lark-<name>
version: 1.0.0
description: "<Skill description and usage scenario>"
metadata:
  requires:
    bins: ["lark-cli"]
---

Body structure:

• ## <Skill Title>
• > **Prerequisites:** Review [../lark-shared/SKILL.md](../lark-shared/SKILL.md).
• ## Commands — bash examples, for example:

# Single-step operation
lark-cli api POST /open-apis/xxx --data '{...}'

# Multi-step orchestration: explain data passing between steps
# Step 1: ... (capture returned xxx_id)
# Step 2: use Step 1's xxx_id

• ## Permissions — list required API scopes or permissions here.

This structure ensures that every skill is self-documented, clearly specifies required binaries, and guides users through the commands used to implement the skill’s logic.

4. Example: List Video Conference Rooms

Suppose you want to list Feishu video conference rooms via API. Your skill command might look like this:

lark-cli api GET /open-apis/vc/v1/rooms --params '{"page_size":"50"}'

Include this in your skill’s SKILL.md under the "Commands" section.

5. Multi-Step

Flows

For more sophisticated workflows, explain how outputs from one API call feed into subsequent steps. For example:

## Step 1:

Search for rooms matching "5F"
lark-cli api POST /open-apis/vc/v1/rooms/search --data '{"query":"5F"}'
## Step 2:

Use one of the returned room IDs to fetch details
lark-cli api GET /open-apis/vc/v1/rooms/<room_id>

Document data dependencies explicitly to ensure clarity and reusability.

When to Use Skill Maker

Skill Maker is the preferred method whenever you need to:

• Turn Feishu API operations into reusable, command-driven skills
• Document and share CLI-based automation with your team or AI agents
• Orchestrate multi-step business processes using Feishu APIs
• Standardize how your organization interacts with Feishu via CLI
• Abstract away API details for non-technical users or automated systems

Use it for both simple atomic API wrappers and more complex, stateful workflows.

Important Notes

• Skills created with Skill Maker require the lark-cli binary available in the runtime environment.
• Always verify API permissions and scopes required for your skill. Document these explicitly in the "Permissions" section of your SKILL.md.
• If you cannot find a direct CLI or shortcut command for your need, fall back to OpenAPI exploration and raw API calls using lark-cli api.
• Follow the recommended structure and documentation practices to ensure your skills are discoverable, maintainable, and easy to use by others or by automation agents.
• Review shared skill dependencies such as lark-shared to ensure compatibility and best practices.

Skill Maker streamlines the process of turning Feishu OpenAPI workflows into shareable, documented CLI skills, empowering both human and AI users to automate and standardize business processes efficiently.