The Agent Skills Directory

[SAFE]: The skill is entirely composed of markdown instructions and reference materials for logical analysis. No malicious patterns, obfuscation, or data exfiltration vectors were detected across the analyzed files.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted long-form text provided by the user (op-eds, manifestos). While this constitutes an attack surface where malicious instructions could be embedded in the input text, the skill's structured workflow and lack of privileged capabilities (such as code execution or network access) render this risk negligible. The output is confined to text-based analysis and Mermaid diagrams.

argument-deconstruct