The Agent Skills Directory

[SAFE]: The skill is strictly instructional and restricted by its configuration to read-only tools (Read, Grep, and Glob), preventing unauthorized system modifications.
[SAFE]: The instructions include strong security safeguards, explicitly prohibiting the agent from exposing API keys or tokens and requiring the use of environment variables for credentials in examples.
[SAFE]: No malicious patterns such as prompt injection, obfuscation, or remote code execution were found. The skill aligns perfectly with its stated purpose as a developer advisor for the Contentstack CMS platform.

cms-taxonomy