voicelayerBLOCK
Voice I/O MCP for AI coding agents — local TTS + STT, 2 tools (`voice_speak`, `voice_ask`) with ToolAnnotations
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
Your AI agent can't hear you and respond to you. VoiceLayer gives it ears and a voice.
[](https://www.npmjs.com/package/voicelayer-mcp) [](https://www.apache.org/licenses/LICENSE-2.0) [](https://modelcontextprotocol.io) [](#the-two-tools)
Voice I/O for AI coding assistants. Press F5, speak to Claude Code, get on-device transcription in under 1.5 seconds. Your AI speaks back. Works with any MCP client.
On a mouse with side buttons, button 5 (forward) works like F5 and button 4 (back) presses Return.
You ──🎤──> whisper.cpp ──> Claude Code ──> edge-tts ──🔊──> You STT (local) MCP tools TTS (free)
Local-first. Free. Open-source. No cloud APIs, no API keys, no data leaves your machine.
[Website](https://voicelayer.etanheyman.com) · [Docs](https://etanhey.github.io/voicelayer/docs/) · [npm](https://www.npmjs.com/package/voicelayer-mcp)
Install
macOS (recommended) — the tap ships the CLI/MCP package and the notarized notch app:
brew tap etanhey/layers brew install etanhey/layers/voicelayer # CLI + MCP server brew install --cask etanhey/layers/voicebar # VoiceBar notch app (owns the mic) voicelayer setup # one-time runtime setup
Install the cask too — the formula alone gives you the CLI/MCP but no notch UI.
npm — if you only want the MCP server:
bun add -g voicelayer-mcp # or: npm i -g voicelayer-mcp brew install sox socat && pip3 install edge-tts brew install whisper-cpp # optional, for local STT # whisper model (~1.5 GB) — needed for local STT mkdir -p ~/.cache/whisper && curl -L -o ~/.cache/whisper/ggml-large-v3-turbo.bin \ https://huggingface.c
1dd0afcd8c52OBSERVED · 2026-10-11Connect
Built from this server's own package name, version and transport as found in its source — not copied from anyone's documentation, so it cannot drift against a page we do not control. Replace the environment placeholders with a token scoped to the least it needs.
claude mcp add voicelayer-mcp --env QA_VOICE_WISPR_KEY=${QA_VOICE_WISPR_KEY} -- npx -y [email protected]Exposed tools (2)
2 read · 0 write · 0 destructive.
| Tool | Risk | Description |
|---|---|---|
voice_ask | read | Ask ONE short question aloud (≤600 chars — longer is refused).\n |
voice_speak | read | Speak an announcement/status aloud or log a note silently. NON-BLOCKING — returns instantly.\n |
Trust audit
BLOCKgrade D · trust 69/100 Do not install this without reading the findings. The audit found something that could harm you or your machine.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | WARN |
| L1 | Static analysis of the code | FAIL |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | PASS |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- declared (6 observation(s))
- Network
- declared (13 observation(s))
- Shell
- declared (4 observation(s))
- Dependencies
- not all pinned
- Secrets in source
- found
Findings (25)
Object.entries(process.env ... curl
VoiceBar.icns
const response = await fetch(`http://127.0.0.1:${port}/health`);const response = await fetch(`http://127.0.0.1:${options.port}/inference`, {const token = "etanhey/layers/voicebar";
clean_speech.wav
high_noise.wav
zero_rms.wav
lead.wav
.claude.local.md.example
.deepsource.toml
.mcp.json.example
.swiftformat
const api = new Function(
const key = createHash("sha1").update(script).digest("hex").slice(0, 16);`chunk-${createHash("sha1").update(chunk).digest("hex").slice(0, 12)}.wav`,import { HybridNotchPrototype } from "../../../components/notch-lab/hybrid/HybridNotchPrototype";import { findDefaultAudio } from "../../scripts/benchmark-stt-decode";"../../docs.local/goldens/pause-aware-boundaries.json",
} from "../../paths";
"../../docs.local/goldens/smart-chunking-clip-b.json",
curl http://127.0.0.1:8880/health \
curl -X POST http://127.0.0.1:8880/synthesize \
export QA_VOICE_STT_POLISH_ENDPOINT=http://127.0.0.1:8080/v1/chat/completions
assert base64.b64decode(response.json()["audio_b64"]) == b"fake-mp3"
Gates applied: no_behavioural_pass.
1dd0afcd8c52full audit observations/trust-audit/mcp-server/etanhey__voicelayer.json · Report an issue / request a re-scanAudit history
Every audit this server has had. A grade with a past is a grade somebody is still checking.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-11 | 1dd0afcd8c52 | BLOCK | D | 69 | first audit |
Questions
What is the voicelayer MCP server?
Voice I/O MCP for AI coding agents — local TTS + STT, 2 tools (`voice_speak`, `voice_ask`) with ToolAnnotations
What tools does voicelayer expose?
2 in total: 2 read-only, 0 that write, and 0 that can delete or overwrite. Every one is listed on this page with its risk.
Is voicelayer safe to connect to an agent?
No — not without reading the findings first. The audit graded it D (69/100) and found 1 critical or high issue in the source. Each one is listed on this page with the file and line it is on.
What credentials does voicelayer need?
It reads QA_VOICE_WISPR_KEY, VOICELAYER_TTS_AUTH_TOKEN_FILE and VOICELAYER_TTS_DAEMON_SECRET_FILE from the environment. Give it a token scoped to the least it needs — an agent that can be talked into calling a tool can be talked into calling it with your credentials.
How does voicelayer run?
It speaks stdio, so it runs as a local process your client starts. It is published on npm as site at 0.1.0.
How current is this page?
The grade is for one exact copy of the source (1dd0afcd8c52), read on 2026-10-11. The repository is watched and re-audited when it changes.