The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes standard macOS and iOS development commands, including find, grep, xcodebuild, plutil, and xcrun simctl. These tools are used for their intended purposes: locating translation catalogs, building the application, and verifying compiled localization resources in the simulator environment.
[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the user's project by scanning Swift source files and .xcstrings catalogs.
Ingestion points: Swift files and JSON-based string catalogs are read by scripts/audit.py to identify translation gaps.
Boundary markers: The skill does not implement explicit delimiters or instructions to ignore potential commands embedded within the strings it translates.
Capability inventory: The skill has the ability to modify project files (.xcstrings) and execute build/utility commands.
Sanitization: There is no explicit sanitization of the processed text for LLM instructions. However, the risk is minimal as the skill lacks network exfiltration capabilities and is scoped strictly to localization management tasks.

swift-missing-translations