The Agent Skills Directory

[SAFE]: The skill functions as a template-based prompt generator. Its primary mechanism involves reading markdown files in the vizprompt/references/ directory to construct JSON responses.
[SAFE]: No high-risk command execution patterns, such as curl, wget, eval, or exec, were found in any of the skill's instruction or reference files.
[SAFE]: The skill does not access sensitive file paths (e.g., credentials, SSH keys, or environment variables) and does not define any network operations that could be used for data exfiltration.
[SAFE]: Obfuscation techniques, including Base64 encoding of instructions, zero-width characters, or homoglyphs, are absent from the provided files.
[SAFE]: The 90+ reference templates are purely data-driven markdown files containing JSON schemas with placeholders for user-supplied arguments. They contain no executable logic or dynamic context injection markers.
[SAFE]: The skill is authored by 'gemini960114' and all internal references and instructions are consistent with its stated purpose of visual prompt generation.

vizprompt