The Agent Skills Directory

[PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection because it is designed to ingest and process untrusted content from external websites during accessibility audits.
Ingestion points: External website content is fetched via Puppeteer and analyzed by Axe-core in resources/implementation-playbook.md.
Boundary markers: The instructions do not currently define explicit delimiters or instructions to ignore embedded commands within the analyzed web content.
Capability inventory: The skill utilizes puppeteer for browser automation and navigation across different file locations and network resources.
Sanitization: No explicit sanitization or filtering of the retrieved web content (such as metadata or text nodes) is performed before the data is presented to the agent.
[EXTERNAL_DOWNLOADS]: The skill's documentation includes examples that reference external dependencies and automated testing workflows.
Fetches official GitHub Actions from the actions organization for CI/CD integration.
Utilizes well-known and established technology packages such as axe-core, puppeteer, pa11y, and jest-axe for its core auditing functionality.

accessibility-compliance-accessibility-audit