The Agent Skills Directory

[SAFE]: The skill is strictly instructional, providing guidelines for WCAG 2.2 and COGA compliance. It does not contain any prompt injection attempts or instructions to bypass safety protocols.
[SAFE]: No network operations (e.g., curl, wget), hardcoded credentials, or exfiltration patterns were identified. The skill is isolated and does not interact with the host system's sensitive data.
[SAFE]: Analysis of the skill body and references found no evidence of obfuscation techniques such as Base64 encoding, zero-width characters, or homoglyphs.
[INDIRECT_PROMPT_INJECTION]: The skill possesses an ingestion surface for untrusted data as it is designed to audit user-provided URLs and content.
Ingestion points: Phase 1 (SKILL.md) requests the user to provide a page URL, component, or content block for review.
Boundary markers: None; the skill does not use specific delimiters or instructions to ignore embedded commands in the source material.
Capability inventory: The skill has no tool access (allowed-tools is not present) and is restricted to outputting a JSON findings report.
Sanitization: No validation or sanitization of the input content is performed before processing.

cognitive-accessibility