The Agent Skills Directory

[SAFE]: The skill provides legitimate guidance for asset localization and media editing using the Layer AI platform tools.
[EXTERNAL_DOWNLOADS]: The skill contains instructions to suggest the installation of sibling skills from the vendor's own repository (layerai/skills) using standard CLI commands. These references target official vendor resources and are safe for the intended use case.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted data such as user-provided translations and media assets.
Ingestion points: External content enters the process through translations and source media file_id parameters.
Boundary markers: The instructions do not specify explicit delimiters or warnings for the agent to ignore potentially malicious instructions embedded within the translation data.
Capability inventory: The skill employs tools for media generation and metadata retrieval, specifically execute_forge, render_text, compose_image, get_base_model, and list_base_models.
Sanitization: There is no mention of sanitizing or filtering external content before it is processed by the media generation tools.

layer-localization