The Agent Skills Directory

[SAFE]: The skill follows security best practices for its intended purpose of site auditing. It correctly identifies the limitations of static fetch tools and recommends secure, official alternatives for tasks like schema validation.
[PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it is designed to ingest and analyze untrusted data from external websites and local marketing context files to perform its audit.
Ingestion points: The skill audits user-provided URLs (via web_fetch or curl) and reads local context files such as .agents/product-marketing.md and .claude/product-marketing.md to tailor its recommendations.
Boundary markers: The instructions do not define specific delimiters or warnings to separate the content retrieved from external websites from the agent's logic.
Capability inventory: The agent's capabilities are restricted to reading files, fetching web content, and generating SEO reports; no high-risk capabilities like arbitrary command execution, persistence, or data exfiltration are present.
Sanitization: No explicit sanitization or filtering of content from the audited websites is mentioned, which is consistent with the skill's primary purpose of diagnostic review.

seo-audit