The Agent Skills Directory

[SAFE]: The skill provides a set of templates and instructions for structuring new skills. No malicious prompt injection patterns or attempts to bypass safety filters were found in the instructional content.
[SAFE]: The included utility script, scripts/validate_skill.py, performs structural checks on markdown files using regular expressions and standard file I/O. It does not utilize dynamic execution or download external content.
[SAFE]: Documentation and templates (e.g., REFERENCE.md, FORMS.md) use progressive disclosure techniques to maintain context efficiency and do not contain hardcoded credentials or sensitive data.
[SAFE]: File operations within the example and template scripts (e.g., analyze.py, process.py) are restricted to reading input files and writing output reports or placeholders within a user-defined directory. No risky system commands or privilege escalation patterns are present.
[SAFE]: All external URL references point to trusted domains, including docs.anthropic.com and github.com/anthropics, which are part of the verified trusted organizations list.

creating-skills