The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill fetches documentation, examples, and component registry items from the official project domain hosted on Vercel (evilcharts-sv.vercel.app). These downloads are integral to the skill's primary function of providing up-to-date API references and components.
[COMMAND_EXECUTION]: Provides instructions to execute npx shadcn-svelte@latest add [URL] to install library components. This command is a standard development workflow for adding source-distributed components to a project.
[REMOTE_CODE_EXECUTION]: The installation process involves downloading component configurations (JSON) from a remote registry which are then used to generate local source code. The source (evilcharts-sv.vercel.app) is the project's primary deployment on a well-known cloud provider.
[INDIRECT_PROMPT_INJECTION]: The skill is configured to ingest and follow instructions from external documentation files (llms.txt, llms-full.txt). While this creates a surface for indirect prompt injection if the remote source were compromised, it is the intended primary operation for keeping the agent's knowledge current.
Ingestion points: External Markdown and text files from evilcharts-sv.vercel.app.
Boundary markers: None explicitly defined in the instructions.
Capability inventory: Command execution (npx), file system writes, and code generation.
Sanitization: Standard LLM guardrails are assumed; no specific skill-level sanitization is documented.

evilcharts-svelte