The Agent Skills Directory

[DYNAMIC_EXECUTION]: The skill uses npx i18next-cli localize --print-agent-prompt to generate its actual operating instructions at runtime. The agent is directed to follow this externally generated output verbatim, effectively creating a dynamic dependency for its core logic.
[REMOTE_CODE_EXECUTION]: The skill suggests installing community-maintained packages, such as i18next-cli-vue (by PBK-B) and i18next-cli-plugin-svelte (by dreamscached), which are hosted in individual user repositories outside the official vendor's organization.
[INDIRECT_PROMPT_INJECTION]: The skill processes the project's entire codebase for string instrumentation and extraction, creating an attack surface for indirect prompt injection.
Ingestion points: The skill reads source files including .js, .ts, .jsx, .tsx, .vue, and .svelte using the i18next-cli instrument and extract commands.
Boundary markers: No explicit markers or warnings to ignore instructions embedded in code are provided for the file-reading phase.
Capability inventory: The agent possesses the capability to modify files and execute shell commands via the Bash tool.
Sanitization: No sanitization of source code content is performed before the data is processed by the agent's tools.
[PERSISTENCE]: The skill instructs the agent to persistently add an MCP server (https://mcp.locize.app) to its configuration using claude mcp add, which maintains access and functionality across future sessions.
[COMMAND_EXECUTION]: The skill utilizes npx to run the i18next-cli and related plugins without pinning them to specific versions, which relies on the security of the latest packages published to the registry at the time of execution.

i18next-localization