The Agent Skills Directory

[NO_CODE]: The skill package is composed entirely of Markdown files (SKILL.md, resources/anti-ai-patterns.md, and resources/translation-rubric.md). It contains no executable scripts (.sh, .py, .js), binaries, or configuration files that could run on the host system.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it is designed to ingest and process untrusted data from the user and the local environment. * Ingestion points: The agent is instructed to read source text for translation and scan project files (including locale files in messages/ or locales/, and surrounding source code) to infer context as described in the Stage 1 and Context Inference sections of SKILL.md. * Boundary markers: There are no explicit delimiters or instructions to the agent to disregard potential commands embedded within the source text or context files. * Capability inventory: No capabilities for network communication, file writing, or command execution are requested or present in the skill's file set. * Sanitization: The skill does not implement any validation or sanitization for the inputs it processes.
[SAFE]: The skill follows secure practices by focusing on a specific linguistic task and lacking access to sensitive system resources or external network endpoints.

oma-translator