The Agent Skills Directory

[SAFE] (SAFE): The file SKILL.md consists entirely of markdown documentation. It serves as a reference guide for developers and contains no executable scripts, system commands, or configuration files that would perform actions on a host system.- [NO_CODE] (INFO): No functional code, dependency manifests (like package.json or requirements.txt), or remote code execution patterns were found. The skill acts as a set of instructions for the AI agent to use as a knowledge base.- [CREDENTIALS_UNSAFE] (SAFE): No credentials or secrets are present. The documentation specifically includes a safety guideline ('args-no-secrets-flags') advising users not to read secrets from command-line flags.- [DATA_EXFILTRATION] (SAFE): There are no network-capable commands or exfiltration vectors. The 'analytics-no-phone-home' rule explicitly discourages unauthorized data transmission.

cli-guidelines