The Agent Skills Directory

[COMMAND_EXECUTION]: The skill explicitly instructs the agent to generate and execute custom scripts (Node.js or TypeScript) to handle bulk operations on translation files. This involves programmatically reading, merging, and writing project files, which is a high-privilege activity.
[COMMAND_EXECUTION]: The instructions suggest using dynamic imports (await import()) to load and process local TypeScript translation files. This technique executes the contents of those files within the agent's runtime environment.
[COMMAND_EXECUTION]: The skill recommends using ts-morph for AST (Abstract Syntax Tree) manipulation. This implies the agent will perform complex code modifications on project source files using external library logic.
[PROMPT_INJECTION]: The skill identifies a surface for indirect prompt injection because it reads and processes content from translation files (JSON, YAML, TS) that could be controlled by an external party.
Ingestion points: Local translation files in the project directory (e.g., en.json, translations.ts).
Boundary markers: None specified for the content of the translation files being processed.
Capability inventory: File system read/write, script generation, and dynamic script execution.
Sanitization: No explicit sanitization or validation of the content within translation files is performed before it is used in scripts or subsequent prompts.
[EXTERNAL_DOWNLOADS]: The skill's documentation references external resources including the sorawit-w/agent-skills marketplace and the ui-ux-pro-max-skill repository on GitHub for additional functionality.

i18n