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 instructions/shreyaskarnik/voice-mcp/claude-mdgit clone --depth 1 https://github.com/shreyaskarnik/voice-mcpWrote 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/instructions/shreyaskarnik/voice-mcp/claude-md)<a href="https://agentmods.dev/instructions/shreyaskarnik/voice-mcp/claude-md"><img src="https://agentmods.dev/badge/instructions/shreyaskarnik/voice-mcp/claude-md.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 5.1 | $0.00614 | $0.00614 |
| Opus 5 | $0.00307 | $0.00307 |
| Sonnet 5 | $0.00123 | $0.00123 |
| Haiku 4.5 | $0.00061 | $0.00061 |
Grade A, and why
voice-mcp CLAUDE.md 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 6d 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 — 43 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.md
This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.
Project Overview
voice-mcp is a bidirectional voice MCP server for Claude Code (and other MCP-compatible agents). It exposes two tools — listen() (STT) and speak() (TTS) — that run locally on Apple Silicon via mlx-audio. The entire server is a single file: server.py.
Commands
uv sync # Install dependencies
uv run server.py # Start the MCP server (stdio transport)
There is no test suite, linter, or build step.
Architecture
Single-file server (server.py, ~273 lines) built on FastMCP (from the mcp SDK). Communicates over stdio — stdout is the MCP JSON-RPC channel, so all logging goes to stderr.
Key components:
- Lifespan hook — pre-loads STT and TTS models at startup so first tool call is fast
- Lazy singletons —
get_stt_model()/get_tts_model()load models once from HuggingFace MLX Community repos - VAD recording —
record_until_silence()uses webrtcvad (mode 3) with energy-based fallback; stops after 1.5s of silence - Audio cues —
chime_listening()/chime_done()synthesize short tones via numpy+sounddevice to signal mic state - TTS playback —
speak()redirects stdout to /dev/null during playback because mlx-audio's AudioPlayer prints to stdout, which would corrupt the MCP stdio channel
Models:
- STT:
mlx-community/Voxtral-Mini-4B-Realtime-2602-int4(16 kHz input) - TTS:
mlx-community/Kokoro-82M-bf16(24 kHz output, 54 voices across 9 languages)
Hooks (.claude/settings.json + .claude/hooks/notify.sh):
- PreToolUse on
mcp__voice__listen— shows macOS notification and forces permission prompt for mic access - PreToolUse on
mcp__voice__speak— shows notification with text preview - PostToolUse on
mcp__voice__listen— shows notification with transcription result
Important Constraints
- stdout is sacred: Never print to stdout in server code. All logging must use
log.*()(configured to stderr). Thespeak()function's stdout redirect pattern exists for this reason. - macOS + Apple Silicon only: mlx-audio requires Metal (Apple GPU). No CPU/CUDA fallback.
- mlx-audio installed from git main: The
mlx-audiodependency is pinned to the GitHub main branch, not PyPI.
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.
- 6d ago First seen · 43 lines · 614 tokens per session scan A 60bc2876c3dc
voice-mcp CLAUDE.md is an instructions file published in the GitHub repository shreyaskarnik/voice-mcp (16 stars, last pushed 6mo ago), licensed Apache-2.0. It adds 614 tokens to every session, about $0.0031 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 2026-08-30.
Other instructions, from other repositories
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
codex AGENTS.md
AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).
spec-kit AGENTS.md
AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.