The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill crawls up to 500 pages of external website content, which represents a large ingestion surface for untrusted data. Malicious content, hidden metadata, or HTML comments on target websites could attempt to influence the agent's audit summary or bypass reporting constraints.
Ingestion points: External HTML content and metadata fetched from URLs provided by the user (as described in SKILL.md).
Boundary markers: The skill does not define specific markers or instructions to isolate crawled data from the agent's primary instructions.
Capability inventory: The skill is capable of writing detailed markdown reports (FULL-AUDIT-REPORT.md, ACTION-PLAN.md), generating a health score, and delegating tasks to multiple sub-agents (seo-technical, seo-content, etc.).
Sanitization: There are no instructions for sanitizing or validating the fetched web content before it is processed by the agent.
[NO_CODE]: The skill instructions in SKILL.md reference scripts/fetch_page.py as a primary tool for fetching HTML content, but this file is missing from the provided skill package.

seo-audit