The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests natural language prompts to generate music, creating a surface for indirect prompt injection if the inputs come from untrusted sources. \n
Ingestion points: The prompt argument used in music.compose and related methods in SKILL.md and references/api_reference.md. \n
Boundary markers: No delimiters or explicit instructions to ignore embedded commands are present in the provided examples. \n
Capability inventory: The skill makes network calls to the ElevenLabs API and writes files to the local system (e.g., output.mp3). \n
Sanitization: No input validation or sanitization is demonstrated for the user-supplied prompts. \n- [EXTERNAL_DOWNLOADS]: The skill depends on official client libraries (elevenlabs and @elevenlabs/elevenlabs-js) which are downloaded from public package registries. \n- [COMMAND_EXECUTION]: The documentation includes shell commands for package installation and curl examples for direct API interaction.

music