The Agent Skills Directory

[SAFE]: The skill provides structured workflows for auditing and creating other agent skills, incorporating checklists for frontmatter, structure, and content quality.
[SAFE]: Security best practices are explicitly documented, such as instructions in references/api-skill-patterns.md to use shell substitution for API credentials to avoid leaking tokens into the conversational context.
[SAFE]: The skill recommends using isolated environments and standard package managers (like uv run) for scripts, ensuring reliable and verifiable execution.
[SAFE]: The use of XML tags for structural boundaries is promoted to improve the agent's ability to parse instructions correctly without ambiguity.
[SAFE]: All external URLs point to legitimate project documentation (agentskills.io), standard libraries, or well-known developer resources.

skill-maker