The Agent Skills Directory

[SAFE]: The skill is entirely instructional and serves as a reference for Web Content Accessibility Guidelines (WCAG). It does not contain any executable scripts, system commands, or configuration that could lead to unauthorized actions.
[PROMPT_INJECTION]: No malicious prompt injection patterns were detected. The instructions clearly define the agent's scope as a teacher and explicitly forbid it from writing or reviewing code, which serves as a functional constraint.
[EXTERNAL_DOWNLOADS]: The skill references several external resources for learning, all of which point to official W3C domains (w3.org) for WCAG specifications and understanding documents. These are trusted, well-known sources for web standards and do not involve automated script execution.
[OBFUSCATION]: The content consists of plain markdown text. No evidence of Base64 encoding, zero-width characters, homoglyphs, or other obfuscation techniques was found.
[DATA_EXFILTRATION]: The skill does not access sensitive local file paths (such as credentials or environment variables) and does not contain any network communication patterns that would lead to data exfiltration.
[DYNAMIC_EXECUTION]: There is no use of dynamic code execution, runtime compilation, or unsafe deserialization. The skill remains strictly within the bounds of a knowledge-based assistant.

wcag-guide