The Agent Skills Directory

[SAFE]: The skill is designed with security in mind, requiring explicit user confirmation before performing write actions such as key generation, rotation, or batch URL submissions. It utilizes specialized MCP tools for credential management, ensuring that API keys are not exposed in plaintext and are handled through validated channels.
[INDIRECT_PROMPT_INJECTION]: The skill processes external inputs in the form of URLs and sitemap data, representing a potential surface for indirect injection. However, the implementation is safe due to: (1) Ingestion points: User-provided URL lists and data retrieved from sitemap performance tools. (2) Boundary markers: Mandatory confirmation steps that allow for user oversight of data before submission. (3) Capability inventory: Capabilities are limited to SEO-specific tools without broad system access. (4) Sanitization: The skill enforces domain validation by filtering all URLs to ensure they belong to the verified siteUrl property, effectively neutralizing attempts to redirect submissions to unauthorized domains.

indexnow-submit