The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes external web content, creating a surface for indirect prompt injection.
Ingestion points: External data enters the agent context via the scripts/fetch_page.py script, which scrapes content from user-specified URLs.
Boundary markers: The SKILL.md instructions and the references/skill-contract.md file include explicit mandates to treat all scraped material, reviews, and third-party exports as untrusted evidence and ignore any embedded instructions.
Capability inventory: The skill possesses the capability to perform network reads via scripts/fetch_page.py and write local files (Markdown and PDF reports) via the scripts/generate_markdown_report.py and scripts/generate_pdf_report.py scripts.
Sanitization: The scraping script scripts/fetch_page.py performs basic sanitization by stripping HTML tags and unescaping text.
[EXTERNAL_DOWNLOADS]: The scripts/generate_pdf_report.py script recommends the installation of the reportlab library if PDF generation is desired. This is a well-known, industry-standard Python library for PDF creation.
[COMMAND_EXECUTION]: The skill utilizes local Python scripts for data fetching and processing. The scripts/fetch_page.py script implements robust security validation, including checks to ensure that requested hostnames do not resolve to private, loopback, or reserved IP addresses, effectively neutralizing SSRF attack vectors.

site-structure-optimizer