The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill documentation refers to the installation of official libraries including the elevenlabs Python package and the @elevenlabs/elevenlabs-js Node.js package from their respective registries. It also includes examples of interacting with the official ElevenLabs API endpoint at https://api.elevenlabs.io. These are expected components for the skill's stated purpose.- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted input through a text-to-audio conversion pipeline. 1. Ingestion points: User-provided strings are passed to the text parameter in the convert method within SKILL.md. 2. Boundary markers: The skill does not implement delimiters or specific instructions to ignore embedded content within the prompt. 3. Capability inventory: The skill uses standard file writing operations to save output to the local disk, as seen in the Python and JavaScript examples in SKILL.md. 4. Sanitization: No explicit sanitization is performed on the user-supplied text before transmission to the remote service.

sound-effects