The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes output from external CLI commands which serves as an ingestion point for untrusted data.\n
Ingestion points: The scripts/audit-cli-help.sh script captures stdout and stderr from target CLI commands (e.g., help text and version info) in the TOP_HELP and sub_help variables.\n
Boundary markers: The script does not use specific delimiters or protective instructions when presenting the captured CLI output to the agent in the final report.\n
Capability inventory: The skill has the capability to execute system commands via bundled shell scripts.\n
Sanitization: No sanitization or filtering is applied to the captured text to prevent embedded instructions from influencing agent behavior.\n- [COMMAND_EXECUTION]: The bundled scripts execute system commands to perform audits.\n
Evidence: scripts/audit-cli-help.sh uses a run_capture function to execute user-specified CLI commands with --help and --version flags.\n- [EXTERNAL_DOWNLOADS]: The skill installation instructions reference external resources from the author's repository.\n
Evidence: README.md provides commands to download the skill and related packs from github.com/yigitkonur/skills-by-yigitkonur.

audit-agentic-cli