The Agent Skills Directory

[SAFE]: The skill is purely instructional, providing guidance on how to structure, write, and test AI agent skills according to progressive disclosure principles.
[SAFE]: External tools referenced (claude-skills-cli, mcpick) and their respective GitHub repositories are resources provided by the skill's author (spences10) for legitimate development tasks like validation and plugin management.
[SAFE]: Documentation includes links to official Anthropic resources and repositories, which are used neutrally to establish best practices.
[SAFE]: No evidence of prompt injection, unauthorized data access, obfuscation, or malicious command execution was found in any of the skill files.

skill-creator