The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill references the use of npx is-agentic, a tool developed by Vercel for auditing website agent-readiness. This involves downloading and executing a package from a well-known registry. Other mentioned monitoring tools (e.g., Peec AI, ZipTie, Semrush) are standard industry utilities.
[INDIRECT_PROMPT_INJECTION]: The skill has a defined surface for processing external and local data, which is necessary for its primary function of SEO auditing.
Ingestion points: Processes website content during audits and reads local context files such as .agents/product-marketing.md and .claude/product-marketing.md to inform recommendations.
Boundary markers: The skill encourages using structured formats (JSON-LD, Markdown frontmatter) and semantic HTML tags, which serve as natural delimiters for processed content.
Capability inventory: No internal scripts for code execution or network operations are included; the skill relies on instructing the agent to use external tools for analysis.
Sanitization: The skill emphasizes structural clarity and semantic correctness rather than specific data filtering, relying on the agent's native safety protocols when handling ingested text.
[SAFE]: The skill is strictly instructional and follows best practices for search engine optimization. It contains no hidden instructions, base64 obfuscation, persistence mechanisms, or attempts to access sensitive system paths (like SSH keys or AWS credentials).

ai-seo