The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill fetches web content and configuration from well-known and trusted sources as part of its primary functionality.
It utilizes r.jina.ai (a well-known service) as a fallback mechanism to convert web pages into clean Markdown.
It integrates with the GitHub API (api.github.com) and raw content server (raw.githubusercontent.com) to retrieve repository READMEs and file contents.
[COMMAND_EXECUTION]: The skill instructions include a set of recommended CLI commands for the user or agent to verify and audit the generated research packs.
Examples include using find, rg (ripgrep), and sed to check the integrity and structure of the output directory.
[DATA_EXFILTRATION]: The skill performs legitimate data collection by scraping content from user-provided URLs and saving it to a local directory.
The default output path is configured to a specific local directory (/Users/zz/Library/Mobile Documents/iCloud~md~obsidian/Documents/zhangAI/Clippings/Reading/), which is consistent with the skill's stated purpose as a material pack collector.
[SAFE]: The skill implements a dedicated sanitization function (_clean_soup) that strips potentially dangerous or irrelevant HTML elements such as <script>, <style>, <form>, and <nav> from the scraped content before saving it to the local system.

1-web-pack