The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from external websites via web_fetch and browser-based rendering tools during SEO audits, creating an indirect prompt injection surface. Ingestion points: Data from audited URLs enters the agent context through fetching tools like web_fetch or browser-based scripting. Boundary markers: The instructions do not specify explicit delimiters or "ignore instructions" tags for external content, though the agent is directed to adhere to a structured audit framework. Capability inventory: The skill's capabilities are limited to analyzing text and generating reports; it lacks write-access to the filesystem or the ability to execute system commands. Sanitization: The skill documentation notes that web_fetch strips <script> tags during conversion, providing a measure of input sanitization.
[DATA_EXFILTRATION]: The skill utilizes network-enabled tools such as web_fetch and curl to retrieve data from user-specified external URLs for SEO analysis. These operations are intrinsic to the skill's primary purpose and do not involve sensitive local file access or exfiltration to unauthorized endpoints.

marketing-seo-audit