The Agent Skills Directory

[SAFE]: The skill consists entirely of natural language instructions and linguistic rules for translation. It does not contain executable scripts, shell commands, or network operations.
[SAFE]: No obfuscation, prompt injection, or data exfiltration patterns were identified. The content focuses on grammatical cases, verb aspects, and terminology specific to the Croatian language.
[SAFE]: The skill correctly identifies standard security practices by instructing the agent to treat identifiers (URLs, tokens, APIs) as Latin-only and preserve them exactly as they appear in the source text.

localize-hr