The Agent Skills Directory

[SAFE]: The skill is composed entirely of markdown-based instructions for the AI agent. It does not include any scripts, binaries, or external configuration files that could be executed.
[NO_CODE]: No executable code blocks or shell command patterns were detected within the documentation. The skill functions solely as a logical framework for accessibility testing workflows.
[INDIRECT_PROMPT_INJECTION]: The skill is intended to process external data such as UI code and interaction flows. While these ingestion points (UI changes, component behavior) represent a surface for indirect prompt injection, the skill's environment is restricted to a read-only sandbox with no dangerous capabilities like network access or file-system writing, effectively neutralizing the risk.

accessibility-tester