The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill fetches and processes content from external websites, creating a surface for indirect prompt injection.
Ingestion points: scripts/fetch_page.py retrieves content from arbitrary URLs.
Boundary markers: SKILL.md and references/skill-contract.md explicitly instruct the agent to treat external content as untrusted evidence and ignore embedded instructions.
Capability inventory: The skill can read local files and write reports but lacks high-risk capabilities like arbitrary shell execution.
Sanitization: Basic metadata extraction is performed, but content is not sanitized for potential injection.
[COMMAND_EXECUTION]: Local Python scripts are used for data processing and reporting.
Evidence: scripts/fetch_page.py and scripts/seo_health_score.py are invoked to perform SEO tasks. These scripts include security checks like SSRF validation.
[EXTERNAL_DOWNLOADS]: The skill mentions an external dependency for optional functionality.
Evidence: scripts/generate_pdf_report.py suggests installing the reportlab package for PDF output.

seo-sitemap