The Agent Skills Directory

[DATA_EXPOSURE_AND_EXFILTRATION]: The skill implements secure secret management by retrieving API keys from environment variables or local .env files. The helper script contains logic to search parent directories for these files, which is a standard practice for development tools and avoids the risk of hardcoded credentials.
[EXTERNAL_DOWNLOADS]: The skill performs network operations exclusively with the official ElevenLabs API (api.elevenlabs.io). These operations are necessary for the skill's primary function of listing voices and generating speech audio.
[INDIRECT_PROMPT_INJECTION]: The skill accepts text input from local files, standard input, or command-line arguments. Although it lacks explicit delimiters to isolate this data, the processed text is used strictly as a payload for an external speech synthesis API and is not interpreted as instructions or code by the agent or the script.
[DYNAMIC_EXECUTION]: The helper script uses only Python standard libraries and does not involve any dynamic code evaluation, runtime compilation, or unsafe deserialization. File system access is limited to reading configuration and writing the resulting audio files.

elevenlabs-tts