The Agent Skills Directory

[SAFE]: The skill serves as a documentation and workflow guide for creating bash scripts. It includes a template that uses defensive programming patterns (e.g., set -euo pipefail, readonly variables) and standard logging practices.
[PROMPT_INJECTION]: The instructions are focused on guiding the agent through development phases. No evidence of instructions designed to bypass safety filters, override system prompts, or extract internal configuration was detected.
[DATA_EXFILTRATION]: The skill does not perform network operations or access sensitive file paths. The provided template does not include hardcoded credentials or mechanisms for data exfiltration.
[COMMAND_EXECUTION]: The bash template contains benign, standard utility commands (basename, pwd, date) for script identification and timestamped logging. There are no dangerous commands, privilege escalation attempts (such as sudo), or persistence mechanisms.
[EXTERNAL_DOWNLOADS]: No remote scripts or packages are downloaded or executed. The workflow references other skills (e.g., @bash-pro, @shellcheck-configuration) which is standard for platform-based workflow orchestration.

bash-scripting