The Agent Skills Directory

[Prompt Injection] (SAFE): No instructions to bypass safety filters or override system prompts were found. The language is purely instructional for the task of accessibility review.- [Data Exposure & Exfiltration] (SAFE): No access to sensitive files or network exfiltration patterns are present. The skill only analyzes provided UI code.- [Remote Code Execution] (SAFE): There are no patterns for downloading or executing remote code. No external scripts are referenced.- [Command Execution] (SAFE): The skill does not use any subprocess, shell, or system command execution capabilities.- [Obfuscation] (SAFE): The content is entirely plain text without any encoding, zero-width characters, or hidden payloads.- [Indirect Prompt Injection] (LOW): The skill reviews external files, which is an ingestion point for untrusted data. However, the skill has no dangerous capabilities (like file writing or network access) that could be exploited by malicious content within those files.

fixing-accessibility