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.
git clone --depth 1 https://github.com/T0mSIlver/localvoxtralWrote 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/plugins/t0msilver/localvoxtral/localvoxtral-remote)<a href="https://agentmods.dev/plugins/t0msilver/localvoxtral/localvoxtral-remote"><img src="https://agentmods.dev/badge/plugins/t0msilver/localvoxtral/localvoxtral-remote/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/plugins/t0msilver/localvoxtral/localvoxtral-remote"><img src="https://agentmods.dev/badge/plugins/t0msilver/localvoxtral/localvoxtral-remote.svg" alt="Reviewed on agentmods" width="80" height="20"></a>Grade A, and why
localvoxtral-remote scanned grade A with 1 finding 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 16d 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
"description": "Publishes Claude Code session context from a REMOTE host to localvoxtral on your Mac, over an SSH RemoteForward tunnel. Command hooks running a bundled POSIX-sh shim that POSTs via curl — needs only sh an What it actually says
{
"name": "localvoxtral-remote",
"description": "Publishes Claude Code session context from a REMOTE host to localvoxtral on your Mac, over an SSH RemoteForward tunnel. Command hooks running a bundled POSIX-sh shim that POSTs via curl — needs only sh and curl on the remote host, and fails open silently when either is missing.",
"version": "1.45.0",
"author": {
"name": "localvoxtral"
},
"homepage": "https://github.com/T0mSIlver/localvoxtral",
"license": "MIT",
"keywords": [
"dictation",
"context",
"voice",
"ssh",
"remote"
],
"userConfig": {
"token": {
"type": "string",
"sensitive": true,
"title": "localvoxtral host token",
"description": "Per-host token issued by localvoxtral on your Mac (Enroll a remote host). It authorizes remote context only and can never make localvoxtral read a local file. Revoke or rotate it in the app at any time.",
"default": ""
},
"port": {
"type": "string",
"sensitive": false,
"title": "Tunnel port on this host",
"description": "Loopback port on THIS host that localvoxtral's SSH RemoteForward binds — the app allocates one per Mac so two Macs enrolled here can never fight over the same bind. Copy the value from the enrollment plan; leaving it empty falls back to the legacy shared port 8473.",
"default": ""
},
"channel_key": {
"type": "string",
"sensitive": true,
"title": "localvoxtral channel key",
"description": "Per-host key localvoxtral's setup stores here. It signs what this plugin's mod and the app send each other over the tunnel, so a process that takes the tunnel's port cannot put text into your prompt. Run setup again in the app to set it.",
"default": ""
}
},
"types": "./types/index.d.ts"
}
What it installs
The manifest is a name and a version. 1 skill travel with it, and installing the plugin installs all of them — 59 tokens a session between them. Each is measured on its own page, and each can be installed alone.
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.
- 16d ago Changed 278315af0175
- 17d ago Changed · +8 lines 182a108e71ef
- 18d ago Changed ea401d0379c4
- 19d ago Changed 226617b931b4
- 20d ago Changed 4f8fcf45b9ff
- 22d ago Changed dba2b986300e
- 23d ago Changed cd094e0f293e
- 24d ago Changed 855f3b4667bd
- 25d ago Changed 0c60dad13fd2
- 29d ago First seen · 34 lines scan A d86e1a05de8f
localvoxtral-remote is a plugin published in the GitHub repository T0mSIlver/localvoxtral (59 stars, last pushed yesterday), licensed MIT. Its token cost is not measured: this kind of file is read by the harness, not the model. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other plugins, from other repositories
cadence-code
Configurable local speech and transcription for natural Claude Code conversations on Apple Silicon.
toolpiper
Connects Claude Code to ToolPiper, the on-device AI platform for macOS: local LLM inference, speech to text, vision and OCR, browser automation, video editing, and system control through one loopback MCP server.
mlx-vlm-skills
Skills for MLX-VLM: command-line and server inference, convert/quantize, adding new models, benchmarking, contributing, cached-model listing, and reproducible GitHub issues.
memo
Local MCP memory for AI agents — MLX on Apple Silicon or CPU sentence-transformers on Linux/Ubuntu, sqlite-vec store, markdown-on-disk in your Obsidian vault. Zero Ollama, zero cloud APIs.
claude-mlx-tts
Voice-cloned TTS notifications using MLX Chatterbox Turbo. Clone any voice from a 20-second sample.
humanizer
Rewrite AI-sounding text so it reads like the writer without changing what it says.