The Agent Skills Directory

[SAFE]: The skill is a documentation and tooling resource for developers building Claude Code skills. It does not contain any instructions that attempt to bypass safety filters, acquire unauthorized persistence, or exfiltrate sensitive data.
[COMMAND_EXECUTION]: Includes a Node.js utility scripts/validate-skill.js intended for local use to verify skill formatting and naming conventions. The script uses standard filesystem modules (fs, path) for read-only checks and does not employ dangerous functions like eval() or exec().
[EXTERNAL_DOWNLOADS]: Contains references to official Anthropic documentation and public GitHub repositories. These are well-known resources and do not initiate any suspicious or unauthorized remote code execution.

creating-claude-skills