The Agent Skills Directory

[SAFE]: No security issues were detected in this skill. The content consists entirely of instructional markdown files and a local shell script used for file structure validation.
[COMMAND_EXECUTION]: The file scripts/validate-agent.sh is a shell script that performs local validation of agent metadata. It uses standard utilities like grep, sed, and awk to check for required fields and character counts. This script is intended for use in a developer environment and does not execute arbitrary or dangerous commands.
[EXTERNAL_DOWNLOADS]: The skill does not perform any network operations or download external content. All reference materials and examples are contained within the local folder structure.
[PROMPT_INJECTION]: The skill provides guidance and templates for constructing system prompts for other agents (meta-prompting). These instructions follow standard agent design patterns and do not include attempts to bypass safety filters, extract system prompts, or override agent constraints in a malicious way.
[DATA_EXFILTRATION]: There are no patterns suggesting data exfiltration. The skill uses tools like Read, Grep, and Glob solely for documentation and local validation purposes.

agent-development