The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: Fetches the installation script for Bun from its official domain (bun.sh) and executes it using bash to set up the PDF conversion environment.
[COMMAND_EXECUTION]:
Orchestrates a multi-step crawling workflow by executing local Python scripts (crawl_item.py, collect_urls.py, etc.) via subprocess.run.
Automatically launches Chromium or Google Chrome instances in remote debugging mode using subprocess.Popen to enable browser automation.
Invokes the bunx command-line utility to transform Markdown files into PDF documents using the mdpdf package.
[INDIRECT_PROMPT_INJECTION]:
Ingestion points: The skill ingests untrusted HTML content and metadata from Zhihu articles and answers via scripts/crawl_item.py and scripts/collect_urls.py.
Boundary markers: No specific boundary markers or instructions to ignore embedded content are used when processing the external data.
Capability inventory: The skill has permissions to write files to the local filesystem and execute shell commands through multiple scripts in the scripts/ directory.
Sanitization: Employs BeautifulSoup for HTML parsing and html2text for Markdown conversion, which provides a layer of sanitization by stripping executable scripts from the source HTML.

zhihu-crawler