The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill requires the installation of the 'next-intl' package via npm. This is a standard and widely-used library for internationalization in the Next.js ecosystem.
[DYNAMIC_EXECUTION]: The configuration in i18n/request.ts utilizes a dynamic import() statement to load translation files based on a locale variable. While this involves loading code from a computed path, it is a documented and standard pattern for the 'next-intl' library to facilitate localized message loading.
[INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to process and generate content based on strings stored in external JSON files (messages/en.json). This represents a potential surface for indirect prompt injection if the translation files were to contain malicious instructions.
Ingestion points: Translation strings are ingested from messages/*.json files.
Boundary markers: None explicitly defined for the JSON values.
Capability inventory: The strings are rendered within React components and metadata.
Sanitization: Standard React rendering provides protection against XSS, but no specific agent-level sanitization for the JSON content is described.

i18n