Notre avis
Installe le moteur de synthèse vocale Kokoro TTS sur les Mac Apple Silicon (M1+).
Points forts
- Vérifications préalables automatiques (architecture, dépendances)
- Utilisation d’uv pour un environnement Python isolé
- Téléchargement du modèle MLX et vérification par synthèse
- Scripts d’installation et de santé complets
Limites
- Exige un Mac Apple Silicon (Intel/Linux non supporté)
- Dépend de la disponibilité du modèle sur HuggingFace
- Nécessite uv et Python 3.13 spécifiquement
Idéal pour configurer rapidement Kokoro TTS sur un Mac M1/M2/M3 avec MLX.
À éviter sur Intel Mac, Linux ou Windows, ou si vous préférez une installation manuelle.
Analyse de sécurité
PrudenceThe skill executes a bash script from a plugin directory using powerful tools (Bash, downloads) for legitimate installation purposes. No destructive or exfiltrating commands are present, but the external script is not reviewed here, and environment variable expansion could be a minor risk.
Aucun point d'attention détecté
Exemples
Install Kokoro TTS engine on this Mac.Set up text-to-speech using Kokoro and MLX on Apple Silicon.name: install description: "Install Kokoro TTS engine on Apple Silicon. TRIGGERS - install kokoro, setup tts, kokoro install, tts setup." allowed-tools: Read, Bash, Glob, AskUserQuestion
Install Kokoro TTS
Install the Kokoro TTS engine: Apple Silicon verification, Python 3.13 venv, MLX-Audio dependencies, model download, and verification synthesis.
Platform: macOS Apple Silicon (M1+) only. Fails fast on Intel/Linux.
Prerequisites
| Component | Required | Check |
| ------------- | -------- | -------------------- |
| Apple Silicon | Yes | uname -m = arm64 |
| uv | Yes | uv --version |
| Python 3.13 | Yes | uv python list |
Workflow
Step 1: Preflight
# Check Apple Silicon
[[ "$(uname -m)" == "arm64" ]] && echo "OK: Apple Silicon" || echo "FAIL: Requires Apple Silicon (M1+)"
# Check uv
command -v uv && echo "OK: uv found" || echo "FAIL: Install with 'brew install uv'"
Step 2: Install
PLUGIN_DIR="${CLAUDE_PLUGIN_ROOT:-$HOME/.claude/plugins/marketplaces/cc-skills/plugins/kokoro-tts}"
bash "$PLUGIN_DIR/scripts/kokoro-install.sh" --install
This performs:
- Verifies Apple Silicon (fails fast on Intel/Linux)
- Creates Python 3.13 venv at
~/.local/share/kokoro/.venvvia uv - Installs MLX-Audio dependencies (mlx-audio, soundfile, numpy)
- Copies
kokoro_common.py,tts_generate.pyfrom plugin bundle - Downloads Kokoro-82M-bf16 MLX model from HuggingFace
- Writes
version.jsonwith mlx_audio version and model ID - Runs verification synthesis ("Warm up.")
Step 3: Verify
PLUGIN_DIR="${CLAUDE_PLUGIN_ROOT:-$HOME/.claude/plugins/marketplaces/cc-skills/plugins/kokoro-tts}"
bash "$PLUGIN_DIR/scripts/kokoro-install.sh" --health
All 6 checks should pass. Print "Installation complete — run /kokoro-tts:health to verify".
Troubleshooting
| Issue | Cause | Solution |
| ------------------- | -------------------- | ------------------------------------ |
| Not Apple Silicon | Intel Mac or Linux | MLX-Audio requires M1+ Mac |
| uv not found | Not installed | brew install uv |
| Model download slow | Large first download | Wait for HuggingFace download |
| Permission denied | Script not +x | chmod +x scripts/kokoro-install.sh |
| Venv already exists | Previous install | Run --uninstall then --install |
Expert Next.js App Router
Developpement
Un skill qui transforme Claude en expert Next.js App Router.
Générateur de README
Developpement
Crée des README.md professionnels et complets pour vos projets.
Rédacteur de Documentation API
Developpement
Génère de la documentation API complète au format OpenAPI/Swagger.