The Agent Skills Directory

[SAFE]: The skill's primary function is to generate static documentation files (e.g., CLAUDE.md) by scanning repository configuration files. It does not perform any network operations or execute external code.
[SAFE]: The instructions include a robust 'Security: Data Boundaries' section that mandates treating all repository content as untrusted data. It specifically forbids interpreting free-text as instructions and prohibits the execution of code found within the repository during analysis.
[SAFE]: The skill uses factual, non-obfuscated language and relies on well-known, trusted sources (such as Arxiv for research context) to inform its logic. No hidden or suspicious triggers were identified in the metadata or the reference files.

agentmd