The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill provides instructions to download and install components using npx shadcn@latest add @codedvisuals/{name}, which executes code from the vendor's private registry hosted at codedvisuals.com. While this is the intended purpose of the skill, it involves executing code from a remote server.
[EXTERNAL_DOWNLOADS]: Component definitions and source code are fetched from https://codedvisuals.com/r/{name}.json during the installation process.
[COMMAND_EXECUTION]: The skill requires the execution of various shell commands including npm install for dependencies and npx shadcn for component installation.
[INDIRECT_PROMPT_INJECTION]: The skill interprets user prompts to generate React code, creating a surface where untrusted data could influence the resulting code or agent actions.
Ingestion points: User requests for visual composition types and property overrides.
Boundary markers: The skill does not provide specific delimiters or warnings to ignore instructions embedded in user-provided content.
Capability inventory: The skill utilizes shell command execution (via npx) and file writing to the project's source root.
Sanitization: No explicit sanitization or validation of user-provided data is described.

codedvisuals