tts commands

Every subcommand on the shared CLI. Slash /tts and the tts binary call the same entry point.

tts <subcommand> [args...]

Config lives in ~/.agent-tts/config.json. Playback commands talk to the daemon over loopback TCP.

Enablement

Command	Effect
`tts on`	Enable automatic speech after completed turns
`tts off`	Disable, stop playback, clear the queue

Command	Effect
`tts summary`	First prose paragraph (default)
`tts closing`	Last prose paragraph
`tts brief`	First sentence of the lead paragraph
`tts full`	Entire cleaned response

Command	Effect
`tts stop`	Interrupt current playback and clear queued jobs
`tts skip`	Skip the current utterance group, continue this response
`tts pause`	Hold this response after cutting the current group
`tts resume`	Continue a paused response
`tts replay [mode]`	Replay the latest turn for this working directory
`tts preview`	Print the utterance queue without speaking

replay accepts summary, closing, brief, or full. Without an argument it uses the configured mode.

Command	Effect
`tts voice prose <name>`	Prose voice
`tts voice header <name>`	Heading / blockquote voice
`tts wpm <n>`	Rate in words per minute
`tts voices`	List voices for the active backend
`tts backend <auto\|macos\|windows\|linux\|cloud>`	Speech engine
`tts cloud provider <elevenlabs\|openai\|azure>`	Cloud vendor
`tts cloud voice <id>`	Provider voice id
`tts cloud key <api-key>`	Stored key (env var preferred)
`tts cloud region <region>`	Azure region

Prefer ELEVENLABS_API_KEY, OPENAI_API_KEY, or AZURE_SPEECH_KEY over tts cloud key.

Command	Effect
`tts status`	Print enabled, mode, voices, backend, data dir
`tts migrate`	Copy legacy `~/.claude/claude-code-tts` into `~/.agent-tts`