The Agent Skills Directory

[SAFE]: The skill serves as an educational resource and a static analysis tool for internationalization workflows.
[COMMAND_EXECUTION]: The skill includes a Python script scripts/i18n_checker.py intended for auditing local source code. The script uses standard library modules (pathlib, re, json) to identify translation patterns and compare locale files. It does not spawn subshells or execute external commands using unsanitized input.
[INDIRECT_PROMPT_INJECTION]: The audit script reads local project files to detect hardcoded strings. While this ingests external data into the agent's context, the risk is negligible as the script only extracts small text fragments (up to 40 characters) for reporting purposes and lacks capabilities to perform dangerous operations based on that content.
Ingestion points: The check_hardcoded_strings function in scripts/i18n_checker.py reads local code files (.py, .js, .tsx, etc.).
Boundary markers: None (Standard CLI utility output).
Capability inventory: The skill is limited to file system read and search capabilities (Read, Glob, Grep).
Sanitization: Matches are sliced to 40 characters to limit the amount of untrusted text processed.

i18n-localization