The Agent Skills Directory

[SAFE]: No malicious patterns or security vulnerabilities were identified in the skill's instructions, scripts, or reference documentation. The skill operates locally and focuses on accessibility auditing.- [COMMAND_EXECUTION]: The skill includes Python scripts (analyze_component.py, generate_fixes.py) used for analyzing component files (JSX, TSX, Vue, Svelte). These scripts rely on the Python standard library for regex-based text parsing and do not use dangerous functions such as eval(), exec(), or subprocess to execute arbitrary system commands.- [EXTERNAL_DOWNLOADS]: The skill does not perform any network operations or external downloads. All analysis logic and reference materials are contained within the skill folder, and it relies on local imports for shared utility functions.- [DATA_EXFILTRATION]: No data exfiltration vectors were detected. The scripts read local component files for the purpose of identifying accessibility violations but do not transmit any code or findings to external endpoints.- [PROMPT_INJECTION]: The instructions in SKILL.md are focused on the primary purpose of accessibility remediation and do not contain language intended to override agent safety protocols, hijack behavior, or extract system prompts.

accessibility-remediation