Skills Publishing and Team Sharing

A developed Skill must go through packaging, testing, and documentation before it can be distributed to team members or publicly released.

This article introduces the complete process from development completion to release.

Pre-release Checklist

Confirm each item before release to avoid distributing incomplete or defective Skills.

Check Item	Check Content	Reference Section
Complete Functionality	All planned features are implemented and tested	Article 22
Error Handling	Common error scenarios have handling and prompts	Article 16
Accurate Description	Trigger test pass rate ≥ 85%	Article 24
Dependencies Declared	requirements.txt or compatibility field is complete	Article 20
No hardcoded secrets	No sensitive information such as API keys or passwords in scripts	Article 28
Version number updated	The version field in SKILL.md has been updated	Article 21
CHANGELOG updated	This change has been recorded	Article 21
Debugging code removed	No temporary print statements, debug checkpoints, or TODOs	Article 23

Packaging as a .skill File

Before release, package the Skill directory into a.skillfile, which is the standard distributable format.

Example

# Basic packaging command
python -m scripts.package_skill my-skill/

# Specify output path and version number
python -m scripts.package_skill my-skill/ \
--output dist/my-skill-v1.2.0.skill \
--version 1.2.0

# After packaging, verify the file is complete (list files in the package)
python -m scripts.package_skill --list dist/my-skill-v1.2.0.skill

打包完成：dist/my-skill-v1.2.0.skill（共 8 个文件，42KB）
包含：
  SKILL.md
  scripts/process.py
  scripts/requirements.txt
  assets/template.xlsx

Writing User Documentation

Provide complete user instructions at the end of SKILL.md or in a separate README.md.

# 数据分析助手 v1.2.0

## 这个 Skill 能做什么

上传 CSV 或 Excel 文件后，自动完成以下工作：
- 统计各列的行数、均值、最大/最小值、空值比例
- 生成数据质量报告，标注需要清理的问题列
- 输出带统计图表的 Excel 报告文件

## 快速开始

1. 在对话中上传你的数据文件（.csv 或 .xlsx）
2. 说："帮我分析这份数据" 或 "生成统计报告"
3. 等待分析完成，下载生成的 Excel 报告

## 系统要求

- Python 3.8+
- pandas >= 2.0（首次运行时自动安装）

## 常见问题

**Q：文件太大怎么办？**
建议文件不超过 50MB。更大的文件请先筛选或抽样后上传。

**Q：支持哪些编码？**
支持 UTF-8 和 GBK。若文件乱码，请另存为 UTF-8 后重新上传。

Distribution Channels

Channel	Applicable Scenarios	Operation Method
Direct file transfer	Internal sharing within a small team	Send the .skill file, the recipient uploads it to Claude
Git repository	Team collaboration, requires version management	Commit to private/public repository, download from dist/
GitHub Releases	Formal release, public sharing	Create a Release, attach the .skill file
Internal CI/CD	Automated distribution within an enterprise	CI packages and pushes to internal file server

Publishing via GitHub Releases

Example

# Create a Git tag and push
git tag v1.2.0 -m "Release v1.2.0: Added chart generation feature"
git push origin v1.2.0

# Create a Release with GitHub CLI and attach the .skill file
gh release create v1.2.0 \
dist/my-skill-v1.2.0.skill \
--title "My Skill v1.2.0" \
--notes "Added chart generation feature, fixed memory issue in large file processing"

Installation Guide Page Template

If your Skill is intended for multiple users, you can write a concise guide for the installation steps.

## 安装方法

1. 前往 [Release 页面](https://github.com/yourname/skills/releases/latest)
2. 下载 `my-skill-v1.2.0.skill` 文件
3. 打开 Claude.ai → Settings（设置）→ Skills
4. 点击"上传 Skill"，选择下载的 .skill 文件
5. 安装完成后，开始新对话即可使用

安装后，在对话中发送以下内容可以验证安装成功：

> "我上传了一个 CSV 文件，请帮我做数据统计分析"

若 Skill 正常工作，Claude 会按照 Skill 的流程引导你完成分析。

Other Extensions

Skills Advanced

Skills Practical Projects

Skills Publishing and Team Sharing

Pre-release Checklist

Packaging as a .skill File

Example

Writing User Documentation

Distribution Channels

Publishing via GitHub Releases

Example

Installation Guide Page Template