The Agent Skills Directory

[COMMAND_EXECUTION]: The skill's primary function is to execute the rg command for searching and filtering text within a directory structure. It also includes scripts for automated installation.
[EXTERNAL_DOWNLOADS]: The installation scripts (scripts/install.sh, scripts/install.ps1) and setup documentation reference official and well-known package managers (Homebrew, APT, DNF, Pacman, APK) to download the binary. These are recognized as trusted system services.
[DATA_EXPOSURE]: While the skill provides examples for searching for sensitive terms like 'token' or 'localhost' in references/cheatsheet.md, these are presented as legitimate developer use cases for local search and do not involve network exfiltration.
[INDIRECT_PROMPT_INJECTION]: As a tool that processes file contents, the skill inherently possesses an attack surface for indirect prompt injection if it parses files containing malicious instructions.
Ingestion points: Shell output from rg commands executed on local files.
Boundary markers: None identified in the command templates.
Capability inventory: Subprocess execution via shell commands.
Sanitization: None identified; the skill passes raw matched lines to the agent context.

rg