The Agent Skills Directory

[DATA_EXFILTRATION]: The Python script scripts/threejs_audio_asset.py is configured to load environment variables from the user's ~/.env file. This is a standard practice for local secret management to avoid hardcoding API credentials.
[REMOTE_CODE_EXECUTION]: The skill facilitates network communication with the ElevenLabs API (api.elevenlabs.io) for sound generation, text-to-speech, and audio cleaning tasks. ElevenLabs is a well-known service for these AI-driven audio operations. Additionally, the skill provides instructions for the agent to execute shell commands that source user profiles (~/.zprofile, ~/.zshrc) to ensure environment variables are available for authentication.
[PROMPT_INJECTION]: The skill processes user-supplied text for audio generation and voice conversion, which creates a potential surface for indirect prompt injection if the inputs are derived from untrusted third-party sources.
Ingestion points: CLI arguments --prompt and --text passed to the threejs_audio_asset.py script.
Boundary markers: No explicit delimiters are specified in the instructions to isolate user-supplied prompts from agent instructions.
Capability inventory: The agent can execute a local Python script, perform network requests to an external API, and write audio files to the project's asset directory.
Sanitization: The script uses standard URL encoding for specific path parameters, but the core text payload is transmitted to the external API as provided.

threejs-audio-generator