detect_non_localizable
Determine if a string value should be translated or left as-is. Use when filtering localization keys, validating translation files, or building translation task lists.
Security Assessment
About detect_non_localizable
The `detect_non_localizable` skill is designed to analyze string values and determine whether they should be translated or left unchanged. It addresses the common challenge in localization workflows of identifying which text elements require translation and which are inherently non-translatable, such as technical codes, file paths, URLs, or configuration keys. By automating this detection, the skill reduces errors in translation files, ensures consistent handling of non-localizable content, and saves time for localization engineers and developers who manage large volumes of text across applications and platforms.
This skill features a comprehensive set of detection patterns to cover a wide range of non-localizable strings. It checks for empty strings, whitespace-only values, pure numbers, single special characters, URLs, email addresses, file paths, version numbers, API endpoints, UUIDs, hash-like strings, configuration keys in SCREAMING_SNAKE_CASE, template variables, icon identifiers, and CSS/style values. The patterns are evaluated in a specific order from most specific to least specific to minimize false positives. When a string is identified as non-localizable, the skill logs the matching pattern for clarity. It also includes rules for conservative handling, ensuring that ambiguous strings are treated as localizable to avoid missing content that requires translation.
The primary use cases for `detect_non_localizable` include filtering localization keys, validating translation files, and generating translation task lists. It is particularly valuable for software developers, localization engineers, and QA teams who need to maintain accuracy and efficiency in translation workflows. By automatically flagging strings that do not require translation, it allows teams to focus resources on translating meaningful content while preventing technical strings from being misinterpreted by translators. The skill supports various domains, including web development, software configuration, UI/UX design, and internationalization projects, making it a versatile tool in any multilingual software environment.
FAQ
How do I use the detect_non_localizable skill?
You provide a string value as input to the skill, and it returns a boolean indicating whether the string should be left untranslated (true) or translated (false). It also logs which detection pattern matched if the string is non-localizable.
Which types of strings are considered non-localizable?
The skill identifies empty strings, whitespace-only strings, pure numbers, single special characters, URLs, email addresses, file paths, version numbers, API endpoints, UUIDs, hash-like strings, configuration keys in SCREAMING_SNAKE_CASE, template variables, icon identifiers, and CSS/style values as non-localizable.
Can this skill handle mixed strings with text and variables?
Yes, mixed content such as 'Hello {{name}}' or 'Visit https://example.com for details' is considered localizable and will return false, ensuring that meaningful text is translated correctly.
Is there a recommended order for pattern evaluation?
Yes, patterns are checked from most specific to least specific to reduce false positives. It is important to follow the order defined in the skill to ensure accurate detection.
Who should use this skill?
The skill is intended for software developers, localization engineers, and QA teams working on translation and internationalization projects who need to accurately filter out non-translatable content.
Install detect_non_localizable
Quick Setup:
- Copy the skill folder to
.claude/skills/ - Claude will automatically detect and use the skill
Repository
ericfitz/tmi-ux