The Agent Skills Directory

[PROMPT_INJECTION]: The skill is designed to ingest and process untrusted data (source code files or web content) to perform accessibility audits, which introduces an indirect prompt injection surface.
Ingestion points: Processes external data via the file-path or url arguments defined in the skill frontmatter.
Boundary markers: The skill does not explicitly define delimiters or instructions to the agent to disregard instructions embedded within the analyzed content.
Capability inventory: The skill's primary purpose is to "Audit and fix" code, implying capabilities for file reading and potential code modification.
Sanitization: No specific input validation or sanitization routines are mentioned for the content being audited.
[EXTERNAL_DOWNLOADS]: The skill references and provides integration examples for several industry-standard accessibility testing libraries and services.
Includes snippets for @axe-core/playwright, jest-axe, and Cypress-axe integrations.
Recommends the use of well-known auditing tools including Lighthouse, WAVE, and axe DevTools.
[COMMAND_EXECUTION]: The documentation includes automated testing snippets that involve executing test runners (Jest, Playwright, Cypress) which run code to perform DOM analysis.

accessibility-wcag