The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill enables an AI-assisted editing and Git-publishing workflow for Markdown, MDC, and data files. This ingestion of external content constitutes a vulnerability surface where malicious instructions in content could potentially influence agent behavior during visual editing or AI-assisted transformations.
Ingestion points: Reads repository source files including Markdown, MDC, YAML, and JSON into the editor context (documented in references/live-editing.md).
Boundary markers: No specific delimiters or "ignore embedded instructions" warnings are defined in the provided instructions for processing external data.
Capability inventory: The skill possesses the capability to commit drafts to Git and perform media uploads to local or external storage (documented in references/deployment.md and references/live-editing.md).
Sanitization: The editor serializes changes back to standard formats like Markdown and MDC during the publication cycle.
[EXTERNAL_DOWNLOADS]: The skill utilizes npx to install official Nuxt modules such as nuxt-studio, @nuxt/content, and @nuxthub/core. It also references documentation and setup guides from the official nuxt.studio domain, which are recognized as well-known resources in the Nuxt development ecosystem.
[COMMAND_EXECUTION]: The instructions provide standard development commands for the environment, including npx nuxt module add for module installation and nuxt build for preparing SSR-capable deployments.

nuxt-studio