The Agent Skills Directory

[COMMAND_EXECUTION]: The scripts scripts/probe_ripgrep.py and scripts/validate.py execute the rg and sh binaries using subprocess.run. These calls use argument lists rather than shell strings, mitigating command injection risks during validation workflows.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied search patterns. It proactively addresses injection risks by instructing the agent to use the -F flag for literal matches and single quotes to prevent shell expansion of special characters.
[SAFE]: The skill relies on local environment tools and official project documentation. No remote code execution, data exfiltration, or persistence mechanisms were detected.

ripgrep