The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill requires the agent to read and transform source code files, creating a surface for indirect instructions to influence agent behavior.
Ingestion points: The agent is instructed to scan and read JavaScript, TypeScript, and JSX files within the user's project directory (e.g., src/).
Boundary markers: The instructions do not specify any delimiters or safety warnings to ignore potentially malicious instructions embedded in the code being migrated.
Capability inventory: The skill utilizes capabilities to install software packages (npm install), execute developer tools (npx lingui, grep), and write or update local files.
Sanitization: No sanitization or validation of the processed source code is performed before the agent acts upon it.
[EXTERNAL_DOWNLOADS]: The skill involves the installation of several Node.js packages related to the Lingui framework.
The packages (@lingui/core, @lingui/react, @lingui/cli, etc.) trace back to the vendor's own namespace and are standard for the migration process.
[COMMAND_EXECUTION]: The skill uses shell commands to manage dependencies and audit the codebase.
Execution includes npm install for package management, npx lingui extract and npx lingui compile for translation workflow, and grep for identifying residual i18next code. These are standard development tools and commands appropriate for the skill's purpose.

migrate-i18next-to-lingui