| name | anime-tts |
| description | Manage anime TTS + effects — change themes, swap sounds, toggle on/off, adjust volume, configure TTS narration |
| user-invocable | true |
| argument-hint | [status|theme|sound|volume|visuals|tts|setup] [args...] |
| allowed-tools | ["Read","Edit","Bash"] |
anime-tts Plugin Manager
Config file: ${CLAUDE_PLUGIN_ROOT}/config.json
Env file: ${CLAUDE_PLUGIN_ROOT}/.env
Sounds dir: ${CLAUDE_PLUGIN_ROOT}/sounds/
Commands
/anime-tts or /anime-tts status
Read config.json and show current state: theme, enabled, volume, visuals, TTS state, voice name, which sound file is mapped to each event. List available themes and voices.
/anime-tts theme <name>
Switch theme. Read config.json, find the theme in themes, copy its sound mappings into sounds, set theme field. Available themes: default (chiptune synth), anime (CC0 OpenGameArt sounds), silent (visuals only). Save config.
/anime-tts sound <event> <path>
Override a single sound. Events: session_start, prompt_submit, pre_tool, post_tool, stop.
- If
<path> is an absolute path or URL, copy it to sounds/custom/ and update config.
- If
<path> is "default", revert that event to the current theme's default.
- If
<path> is "off", set that event's sound to null.
Show the user how to preview: afplay ~/Coding/anime-tts-plugin/sounds/custom/<file>.wav
/anime-tts volume <0.0-1.0>
Set volume in config.json. Validate range 0.0 to 1.0.
/anime-tts visuals <on|off>
Toggle terminal visual animations (sparkles, beams, etc.) independent of sound.
/anime-tts on / /anime-tts off
Toggle enabled in config. When off, all hooks skip sound, visuals, and TTS.
/anime-tts tts <on|off>
Toggle TTS narration. When on, Claude's response is summarized locally (fast regex extraction — no LLM call, no extra API cost) and spoken aloud via ElevenLabs when Claude finishes responding.
TTS ships disabled by default — users must run /anime-tts tts setup (or manually provide an API key + enable) before hearing anything.
/anime-tts tts setup
First-run onboarding. Walk the user through enabling TTS:
- Check if
ELEVENLABS_API_KEY is set (env var or .env). If missing, prompt: "Paste your ElevenLabs API key (get one at https://elevenlabs.io). I'll write it to .env." Then call /anime-tts tts key <pasted>.
- Ask which voice they want. Show the list from
/anime-tts tts voice. Default suggestion: rei.
- Ask which events should speak. Default:
["stop", "error"] only (quiet). Offer to add session_start and prompt_submit for more chatter.
- Set
tts.enabled = true in config.json.
- Run
/anime-tts tts test to verify.
/anime-tts tts key <api_key>
Write the provided ElevenLabs API key to .env (create the file if missing, update the ELEVENLABS_API_KEY= line if present). Never echo the key back to the user — confirm with Key saved (sk_****...****). If the user passes no key, tell them the three ways to provide one:
- Run
/anime-tts tts key sk_your_key_here
- Add
ELEVENLABS_API_KEY=sk_your_key_here to ~/Coding/anime-tts-plugin/.env
- Export in shell:
export ELEVENLABS_API_KEY=sk_your_key_here
/anime-tts tts events <event1,event2,...>
Set which events trigger TTS. Valid: stop, error, session_start, prompt_submit. Update tts.on_events in config.json.
/anime-tts tts voice <name>
Switch TTS voice. Available voices from config.json tts.voices:
rei — Rei Ayanami, soft breathy Japanese female, quiet monotone (default)sora — Calm Japanese anime male protagonistakira — Charismatic Japanese maleyui — Bright cute Japanese femaleyuki — Youthful calm anime teen femaledylo — Dark fierce young maleitsuki — Energetic tomboyish anime girlaerisita — Kawaii bubbly femininemegumin — Theatrical chuunibyou explosion mageritsuto — Anime prince
Update tts.voice_id in config.json to the selected voice's ID.
/anime-tts tts volume <0.0-1.0>
Set TTS-specific volume in config.json tts.volume.
/anime-tts tts model <model_id>
Switch ElevenLabs model. Options: eleven_flash_v2_5 (fast, default), eleven_multilingual_v2 (higher quality), eleven_turbo_v2_5 (fastest).
/anime-tts tts test [text]
Test TTS by running: echo '<text>' | node ~/Coding/anime-tts-plugin/scripts/test-tts.js
If no text given, use "Mission complete. The code has been modified successfully."
/anime-tts setup dbz
Run the DBZ setup script. Tell the user:
- Download the free pack from https://heltonyan.itch.io/retroanimesfx
- Unzip it
- Then run:
node ~/Coding/anime-tts-plugin/scripts/setup-dbz.js /path/to/unzipped/folder
/anime-tts add-theme <name>
Create a new custom theme. Ask which WAV files to use for each event. Add it to themes in config.json.
Environment keys
ELEVENLABS_API_KEY — stored in .env, required for TTS
Response format
After making changes, show a compact summary table of the new state. Include preview commands for any changed sounds.
