The Agent Skills Directory

[SAFE] (SAFE): The skill consists entirely of documentation and guidelines for applying the Zachman Framework. It does not contain executable code, shell commands, or network operations.
[DATA_EXPOSURE] (SAFE): The skill requires access to standard file reading tools (Read, Glob, Grep) to analyze code architecture, which is consistent with its stated purpose of architectural analysis.
[INDIRECT_PROMPT_INJECTION] (LOW): While the skill ingests untrusted codebase data via Read/Grep tools, it lacks dangerous write-access or network capabilities that would allow for exfiltration or persistence. The primary risk is the agent being misled during analysis by comments in the code, which is inherent to any code-reading tool.

zachman-analysis