The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses ffmpeg to process, compress, and chunk audio files for transcription. It also uses shell commands to load environment variables from a local configuration file. These commands are standard for the described audio processing and development workflow.
[EXTERNAL_DOWNLOADS]: Employs uvx to dynamically run Python tools with dependencies such as httpx, socksio, and openai-whisper. These packages are well-known in the developer community and are used for their intended purposes of network communication and local transcription.
[DATA_EXFILTRATION]: Performs transcription by sending audio data to the OpenAI API endpoint https://api.openai.com/v1/audio/transcriptions. This network activity is documented and necessary for the skill's functionality, with no evidence of data being sent to suspicious or unauthorized domains.
[CREDENTIALS_UNSAFE]: References the storage and loading of API credentials from a ~/code/.env file. This is a recommended practice for secret management, ensuring that sensitive keys are not hardcoded within the instructions or scripts.

wjs-transcribing-audio