The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes several shell commands (primarily grep) to search the project's source code for specific design patterns, accessibility violations, and performance anti-patterns.
[EXTERNAL_DOWNLOADS]: The skill uses npx to download and execute well-known auditing packages from the npm registry, including pa11y, source-map-explorer, and vite-bundle-visualizer.
[INDIRECT_PROMPT_INJECTION]: The skill creates an attack surface for indirect prompt injection by ingesting and processing untrusted data from the local project environment.
Ingestion points: Scans source files (TSX, JSX, CSS, Swift, Kotlin) in the src/ directory and build artifacts in the dist/ directory.
Boundary markers: The instructions do not define explicit boundary markers or delimiters for the ingested code content.
Capability inventory: The skill uses grep for pattern matching and npx for executing external analysis tools.
Sanitization: There is no evidence of sanitization or filtering of the file contents before they are processed by the agent.

design-audit