The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill demonstrates an open redirect pattern in examples/api-routes.md where a URL parameter is used directly in a redirect function.
Ingestion points: In src/routes/api/redirect/+server.ts, the target URL is retrieved directly from url.searchParams.get("url").
Boundary markers: There are no validation checks or domain whitelisting procedures implemented.
Capability inventory: The skill utilizes the redirect(307, target) function to navigate the user to the supplied URL.
Sanitization: No input sanitization or verification against an authorized list of domains is performed.
[INDIRECT_PROMPT_INJECTION]: The skill utilizes potentially unsafe HTML rendering in examples/core.md which could lead to XSS.
Ingestion points: Data is retrieved from the database in src/routes/blog/[slug]/+page.server.ts and passed to the frontend via the load function.
Boundary markers: None are implemented in the code example.
Capability inventory: The {@html data.post.content} expression in src/routes/blog/[slug]/+page.svelte renders raw HTML content.
Sanitization: The example does not demonstrate sanitization of the database content before it is rendered to the user.

web-meta-framework-sveltekit