Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add commands/cris-m/claude_voice/configgit clone --depth 1 https://github.com/cris-m/claude_voiceWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/cris-m/claude_voice/config)<a href="https://agentmods.dev/commands/cris-m/claude_voice/config"><img src="https://agentmods.dev/badge/commands/cris-m/claude_voice/config.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00010 | $0.02137 |
| Opus 5 | $0.00005 | $0.01069 |
| Sonnet 5 | $0.00002 | $0.00427 |
| Haiku 4.5 | $0.00001 | $0.00214 |
Grade A, and why
config scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 5d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 202 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Voice Configuration Wizard
Interactively configure the user's voice settings and save them to voice_config.json.
Instructions for Claude
Walk the user through choosing their voice settings step by step. After each choice, move to the next. At the end, write the config file and test it.
First, read the current config from ${CLAUDE_PLUGIN_ROOT}/voice_config.json to show what is currently set.
Step 1: Ask about provider
Check which provider is currently configured. Ask if they want to change it.
- Kokoro — Lightweight, fast, 54 built-in voices, 8 languages. Good for quick notifications.
- Chatterbox — Higher quality, emotion control (exaggeration, cfg_weight), voice cloning from audio. Best for expressive speech.
- MLX-Audio (macOS Apple Silicon only) — Qwen3-TTS, emotion via text instructions, fastest on M-series chips. Best for natural assistant voice.
If they switch providers, tell them to re-run /claude-voice:init to install the correct dependencies.
Step 2: Language (Kokoro only)
Only ask this if provider is kokoro. Chatterbox and MLX auto-detect language.
Supported Kokoro languages:
| Code | Language | Voices | Needs dependency |
|---|---|---|---|
en-us |
English (American) | 20 | No |
en-gb |
English (British) | 8 | No |
ja |
Japanese | 5 | Yes |
zh |
Mandarin Chinese | 8 | Yes |
es |
Spanish | 3 | Yes |
fr |
French | 1 | Yes |
hi |
Hindi | 4 | Yes |
it |
Italian | 2 | Yes |
pt |
Brazilian Portuguese | 3 | Yes |
Use the simple config codes above in voice_config.json — they are mapped to espeak codes at runtime.
If non-English, install dependencies:
| Language | Install command |
|---|---|
| Japanese | cd "${CLAUDE_PLUGIN_ROOT}" && uv pip install "misaki[ja]" && .venv/bin/python -m unidic download |
| Chinese | cd "${CLAUDE_PLUGIN_ROOT}" && uv pip install "misaki[zh]" |
| Korean | cd "${CLAUDE_PLUGIN_ROOT}" && uv pip install "misaki[ko]" |
| Others | cd "${CLAUDE_PLUGIN_ROOT}" && uv pip install misaki |
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 5d ago First seen · 202 lines · 10 tokens per session scan A ee8f07856b68
config is a command published in the GitHub repository cris-m/claude_voice (4 stars, last pushed 5mo ago), licensed MIT. It adds 10 tokens to every session and 2,137 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.