The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze external web pages, components, and source code. This introduces a surface for indirect prompt injection, where malicious instructions hidden in the audited material (such as in HTML comments or metadata) could attempt to manipulate the auditor's findings or recommendations.
Ingestion points: Web pages, components, and code flows provided for audit as described in SKILL.md.
Boundary markers: Absent. The skill does not instruct the agent to use specific delimiters or to ignore potential instructions within the content being audited.
Capability inventory: The skill generates code fixes and detailed audit reports as seen in references/common-issues-and-fixes.md and templates/audit-report-template.md.
Sanitization: Absent. No instructions are provided for validating or sanitizing the input data before it is processed by the agent.
[EXTERNAL_DOWNLOADS]: The skill suggests the use of various industry-standard libraries and tools for automated accessibility testing and linting.
Evidence: references/testing-tools-and-techniques.md mentions eslint-plugin-jsx-a11y, axe-core, Pa11y, Playwright, Jest, eslint-plugin-vuejs-accessibility, and @angular-eslint. These are well-known and widely used packages within the web development community.

accessibility-audit