The Agent Skills Directory

[SAFE]: The skill provides educational content and architectural templates for agent development. The system prompt patterns and triggering examples align with established AI safety and engineering practices.
[COMMAND_EXECUTION]: The skill includes a local utility script, scripts/validate-agent.sh, which is designed to verify the formatting and frontmatter of agent files. The script uses standard shell commands (grep, sed, awk, head, tail) to perform local text analysis safely and does not exhibit any patterns of unauthorized file access, network communication, or code injection.

agent-identifier