Gandr Mcp

io.github.Gandr-AIPublished by GitHub account Gandr-AIv0.2.0Updated Oct 24, 2026

Sets Gandr up from the live docs, then speaks. 23 languages, six voices, every render watermarked.

VerifiedSTDIODesktop onlyMedia & DesignAI & ML

Overview

AI-generated overview

Gandr's official text-to-speech server that lets an MCP agent render speech in 23 languages with six voices and check token usage.

What you can do
  • Render text to speech as base64 WAV, choosing voice, language, sample rate, pitch and pacing
  • List the six available voices
  • List the 23 supported languages and their codes
  • Check tokens used by the configured API key, where one token equals one character
When to use it
Useful when an agent should produce spoken audio from text — narration, voice replies, or audio assets — and you want it inside Claude Desktop, Cursor or another MCP client without writing API code.
Requirements
Runs as a local process installed from PyPI (pip install gandr-mcp), so Python is needed. A Gandr API key in the GANDR_API_KEY environment variable is required for speech; the setup and listing tools work without it. Network access to Gandr's service is implied.
Before you install
The API key is a secret passed via GANDR_API_KEY and should not be shared. Usage is metered in tokens (one per character) against that key, so long texts consume quota. Every render is watermarked, and the text you synthesize is sent to Gandr's service.

Installation

In SourceWeft

  1. Open Gandr Mcp in the dashboard and add it to a workspace.
  2. Enable the server for the chats that should use its tools.

Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.

Other MCP clients

Follow the launch instructions in the repository.

README

gandr-mcp

The official Gandr MCP server, text to speech inside any MCP-compatible agent (Claude Desktop, Cursor, and friends).

[smithery badge]

Tools

ToolWhat it does
synthesizeRender text to speech (WAV, base64). Voice, language, sample rate 8000-48000, pitch and pacing dials.
list_voicesThe six available voices.
list_languagesThe 23 supported languages and their codes.
get_usageTokens used by the configured key. One token is one character.

Install

bash
pip install gandr-mcp

Claude Desktop (add to claude_desktop_config.json):

json
{  "mcpServers": {    "gandr": {      "command": "gandr-mcp",      "env": {        "GANDR_API_KEY": "gnd_..."      }    }  }}

Get a key at gandr.ai. Docs: gandr.ai/docs.

mcp-name: io.github.Gandr-AI/gandr-mcp

Source: README.md at commit 9df715e

Tools

0
SourceWeft has not started this package in its sandbox yet, so its tools are not known.

Version history

2
  1. v0.2.0LatestOct 15, 2026
  2. v0.1.8Sep 16, 2026

Score

50% of the applicable checks met

How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.

  • Reached in the last 30 days25 pts

    Keep the endpoint answering the MCP handshake; SourceWeft probes it every 30 days.

  • Auth declared10 pts

    Declare the credential headers in the registry entry, or keep the endpoint answering the probe so its auth can be classified.

  • Tools listed15 pts

    Answer tools/list without a credential, or ship a package that starts in a sandbox without one.

  • DescriptionRequired15 pts
  • Logo10 pts
  • Documentation15 pts
  • License10 pts