The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted data from user briefs and owner materials to generate code and documentation, creating a potential attack surface.
Ingestion points: The skill ingests facts about the user's business, audience, and requirements in Step 1 (Brief).
Boundary markers: There are no explicit boundary markers or instructions to isolate or ignore potentially malicious instructions embedded within the user-supplied content.
Capability inventory: The skill possesses the ability to perform network GET requests to an external API (adddomain.net) and generate executable HTML, CSS, and JavaScript frontend code.
Sanitization: No explicit sanitization or validation of the user-provided data is mentioned before it is processed or used in API parameters.
[DATA_EXFILTRATION]: The skill performs network operations to an external domain to provide domain availability and purchasing services.
Evidence: The skill uses GET requests to https://www.adddomain.net/api/* in Section 3 to check domain status and retrieve purchase links. While these operations are documented as core features and target the vendor's own infrastructure (as indicated in the skill metadata), they involve sending user-derived strings to a non-whitelisted external endpoint.

frontend-design-website