The Agent Skills Directory

[SAFE]: This skill functions as a documentation and development resource for building autonomous agents. It contains templates and architectural guidance.
[COMMAND_EXECUTION]: A local shell script scripts/validate-agent.sh is provided to check the structural integrity of agent files. The script uses standard POSIX utilities (grep, sed, awk) for text processing and does not interact with the network or external APIs.
[PROMPT_INJECTION]: The skill includes instructions for generating agent system prompts. These instructions focus on defining expert personas and operational workflows rather than attempting to subvert safety guidelines or override model constraints.
[DATA_EXFILTRATION]: There are no patterns suggesting the collection or transmission of sensitive user data, environment variables, or hardcoded secrets.

agent-identifier