The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to process and modify user-supplied interface code, creating a surface for potential injection attacks through untrusted data.
Ingestion points: Project source files (HTML, CSS, JavaScript) accessed through Read, Edit, Grep, and Glob tools.
Boundary markers: The instructions lack specific delimiting instructions or warnings to ignore commands potentially embedded within analyzed code comments or text.
Capability inventory: The skill possesses the ability to write and edit files and execute shell commands via npm, npx, and node tools.
Sanitization: No input sanitization or filtering logic is present to validate the safety of the ingested content.
[EXTERNAL_DOWNLOADS]: The skill facilitates the use of standard web auditing tools for its primary purpose.
Evidence: Usage of npx @axe-core/cli and npx lighthouse for accessibility testing.
Source: These tools are sourced from the official NPM registry, a well-known and trusted service provider.

accessibility-expert