The Agent Skills Directory

[SAFE]: The skill instructions and reference documents (references/*.md) focus exclusively on providing accessibility audits for web forms according to WCAG standards. No prompt injection markers, role-play bypasses, or instructions to ignore safety guidelines were found.\n- [SAFE]: The skill does not perform any network operations, file system modifications, or subprocess executions. It contains no hardcoded credentials or sensitive file paths.\n- [SAFE]: The configuration in 'SKILL.md' (disable-model-invocation: true) and 'agents/openai.yaml' (allow_implicit_invocation: false) implements security best practices by limiting the agent's autonomous capabilities.\n- [SAFE]: No obfuscation techniques, hidden URLs, or unverifiable external dependencies were identified in the analyzed content.

forms-specialist