The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process content from external websites during SEO audits and research tasks.
Ingestion points: The skill crawls public URLs and extracts text from live pages for auditing and retrievability testing (as seen in references/audit.md and references/validation-evidence.md).
Boundary markers: No specific delimiters or instructions to ignore embedded commands are mandated for the processing of external content.
Capability inventory: The skill utilizes a Python script (scripts/prompt-panel.py) for API interactions and suggests the use of shell probes for validation.
Sanitization: Basic escaping is mentioned for JSON-LD implementation in references/nextjs-implementation.md, but general sanitization of ingested text for report generation is not explicitly detailed.
[COMMAND_EXECUTION]: The skill incorporates a local Python script (scripts/prompt-panel.py) to perform API-based search visibility tests. This is a functional requirement of the skill and uses standard libraries.
[EXTERNAL_DOWNLOADS]: The skill communicates with official API endpoints of well-known service providers (OpenAI, Anthropic, and Google) to measure visibility. These communications are part of the intended SEO monitoring functionality.

seo