The Agent Skills Directory

[SAFE]: The skill provides documentation and best practices for i18n and L10n across various frameworks (React, Next.js, Python).
[COMMAND_EXECUTION]: The bundled script scripts/i18n_checker.py performs read-only operations on a user-specified directory to audit code and locale files. It does not perform network requests, write files, or execute shell commands.
[DYNAMIC_EXECUTION]: No dynamic code execution or unsafe deserialization patterns were detected. The script uses json.loads for parsing locale files and re.findall for string matching.
[INDIRECT_PROMPT_INJECTION]: The script reads content from local project files. While these files are technically untrusted data, the script only performs heuristic regex matching and key counting for reporting purposes, posing no risk to the agent context or execution flow.

i18n-localization