The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze and critique external source code provided by the user, which presents an attack surface for instructions embedded in data.
Ingestion points: The agent is instructed to read source directories (e.g., components/, app/, src/) and analyze files using grep, readFileSync, and specialized auditing tools.
Boundary markers: There are no instructions to treat analyzed code as untrusted or to ignore instructions found within the user's project files.
Capability inventory: The skill has access to powerful shell commands including node, npm, npx, and git.
Sanitization: The skill does not mention sanitizing or escaping the content of the analyzed files before processing them.
[COMMAND_EXECUTION]: The skill relies on several shell-based tools to perform its functions.
Executes a script from a sibling skill at ~/.claude/skills/visant-killer/scripts/killer-scan.mjs using node.
Instructs the agent to run npx impeccable detect for design auditing and npx shadcn@latest add for component management.
Includes standard build and verification commands like npm run build, npx tsc, and pytest.
[EXTERNAL_DOWNLOADS]: The skill references and downloads external resources required for its design system and principles.
Fetches UI component definitions from https://ui.visantlabs.com/r/{name}.json (vendor-owned domain).
References the revenue-centric-design project on GitHub for persuasion mechanics.

visant-frontend