The Agent Skills Directory

[NO_CODE]: The skill consists entirely of Markdown files providing instructions and reference data for SEO/AEO/GEO auditing. It does not include any scripts (Python, Node.js), binaries, or configuration files that execute code.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to audit external web pages or content provided by the user. This creates an attack surface where instructions embedded in the analyzed content could attempt to influence the agent.
Ingestion points: External web pages or copy provided by the user (SKILL.md, AUDIT/CHECK modes).
Boundary markers: The skill uses fixed output formats (RETRIEVAL SPEC) and strict Gates (R1–R8) which act as structural constraints.
Capability inventory: None; the skill is limited to generating text reports and providing analytical advice.
Sanitization: None mentioned, but the skill instructions prioritize critical grading of claims.
[COMMAND_EXECUTION]: The instructions mention using curl to verify server-side rendering of web pages. This is standard practice for SEO auditing and is framed as a manual verification step for the user/agent rather than a hidden or malicious command.

retrieval-architecture