The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill facilitates the installation of UI components from a custom shadcn registry at ui.visantlabs.com using the npx shadcn command. These resources are part of the vendor's ecosystem.
[COMMAND_EXECUTION]: The skill utilizes standard shell commands (grep, find) and executes a bundled Node.js utility (refine-scan.mjs) to scan the target repository for design system tokens and UI anti-patterns. These operations are conducted locally on the developer's machine.
[INDIRECT_PROMPT_INJECTION]: The skill ingests and processes external source code data, creating a potential attack surface for indirect prompt injection.
Ingestion points: Target repository files (primarily .tsx and .jsx) that are read and analyzed by the scripts/refine-scan.mjs script.
Boundary markers: The skill does not explicitly define delimiters or specialized instructions to prevent the agent from processing malicious or deceptive content embedded within the scanned source code files.
Capability inventory: The skill is capable of reading the local file system and executing Node.js scripts and CLI tools for repository analysis.
Sanitization: The linter script uses specific regular expression patterns to identify programmatic UI defects rather than evaluating natural language content, which significantly reduces the risk of the agent obeying instructions embedded in the code.

visant-refine