The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes various shell commands to manage its lifecycle. It uses mkdir and echo to initialize configuration in .listenhub/tts/config.json, and employs cat and jq to read and manipulate JSON settings. It interacts with the listenhub binary to generate audio content, specifically calling listenhub tts create and listenhub creation get for status polling.
[EXTERNAL_DOWNLOADS]: The skill uses curl to download generated audio files (.mp3) from URLs provided by the listenhub service. These downloads are performed to the local working directory after user confirmation.
[INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied text and scripts for conversion to speech.
Ingestion points: Text is ingested via user prompts in SKILL.md (Step 1 in both Quick and Script modes).
Boundary markers: The skill uses a <HARD-GATE> requiring explicit user confirmation of the text and voice selection before any generation command is executed.
Capability inventory: The skill can execute shell commands via the listenhub CLI and download files via curl.
Sanitization: The skill uses jq to safely wrap user input into JSON objects for the CLI, reducing the risk of command injection via the text content.

tts