Claude MarketplacesSubmit

Macwhisper Mcp Server

docdyhr/macwhisper-mcp-server

Connects Claude Desktop to MacWhisper's command-line interface for fully local audio transcription on macOS. Exposes a transcribe_audio tool that accepts file paths, optional model selection, and a persist flag to save transcripts to MacWhisper's history. Also includes list_models to see installed Whisper models, cancel_transcription for long jobs, and folder watching tools that auto-transcribe new audio files. Everything runs on your Mac with no cloud APIs. Requires MacWhisper installed with CLI enabled and an allow-list of directories configured in your Claude Desktop config. Reach for this when you want Claude to transcribe, summarize, or extract action items from audio files without sending anything off your machine.

GitHub stars: 3
tools: 7
auth: Required
transport: STDIO
env vars: 3
last updated: Oct 15, 2026

Tools

7 tools

Public tool metadata for what this MCP can expose to an agent.

transcribe_audioTranscribe an audio file and return the transcript as plain text.

Transcribe an audio file and return the transcript as plain text.

No parameter schema in public metadata yet.

list_modelsList transcription models installed in MacWhisper, plus whisper-cpp models if MACWHISPER_WHISPERCPP_MODEL_DIR is configured; active MacWhisper model is marked

List transcription models installed in MacWhisper, plus whisper-cpp models if MACWHISPER_WHISPERCPP_MODEL_DIR is configured; active MacWhisper model is marked

No parameter schema in public metadata yet.

cancel_transcriptionCancel the currently running transcription

Cancel the currently running transcription

No parameter schema in public metadata yet.

list_allowed_pathsReturn the directories the server is allowed to read from

Return the directories the server is allowed to read from

No parameter schema in public metadata yet.

start_watchWatch a folder and auto-transcribe new audio files into ../done/

Watch a folder and auto-transcribe new audio files into ../done/

No parameter schema in public metadata yet.

stop_watchStop the active folder watcher

Stop the active folder watcher

No parameter schema in public metadata yet.

get_watch_resultsReturn completed watch-folder transcriptions and clear the queue

Return completed watch-folder transcriptions and clear the queue

No parameter schema in public metadata yet.

Sponsors

Promote your product hereReach 125k+ visitors monthly Promote your product hereReach 125k+ visitors monthly

Agent CaffeineSponsoredClose your lid, come back to finished work

SnugSponsoredThe clipboard manager for the AI era Promote your product hereReach 125k+ visitors monthly

WaxSponsoredBlock anything with supervision, without erasing your iPhone Promote your product hereReach 125k+ visitors monthly

knobsSponsoredSee your site on any phone, any day, in any language

Configuration

MACWHISPER_ALLOWED_PATHS*

Colon-separated list of directories the server may read audio files from (e.g. ~/Desktop:~/Downloads)

MACWHISPER_CLI

Path to the mw binary. Auto-detected from /Applications/MacWhisper.app if omitted.

MACWHISPER_LOG_PATH

Log file path. Defaults to ~/Library/Logs/macwhisper-mcp.log

Readme

Deploy on Railway — $20 free credits

macwhisper-mcp-server

Local MCP server that connects MacWhisper to Claude Desktop.

What it does: Drop an audio file on your Desktop, then ask Claude to transcribe it, summarise it, or pull out action items — in one step. MacWhisper does the transcription on your Mac; Claude does the thinking. Nothing leaves your machine. No cloud APIs. No data ever leaves your Mac.

Audio file  →  MacWhisper CLI  →  MCP server  →  Claude Desktop

View the full README on GitHub

Details

Repository: github.com/docdyhr/macwhisper-mcp-server
Category: Media & Entertainment
Package: macwhisper-mcp-server
Transport: STDIO
Auth: Required
Registry: MCP registry, active

Related Media & Entertainment MCP servers

1

Facetimemcpaernouddekker/macos-mcpMCP server for macOS FaceTime — initiate phone, FaceTime audio, and FaceTime video calls0stars

2

Ffvoicechicogong/ffvoice-engineOffline speech-to-text & speaker diarization MCP server: transcribe audio on-device, no cloud0stars

3

CutProgetcutpro/mcpCreate images, videos, music, voices, sound effects and 3D with any AI model, plus AI video clips.0stars

4

Dynamoigetdynamoi/dynamoiPromote a song on Spotify or promote a video on YouTube from your AI agent. Free Smart Links.3stars

5

Barevaluebarevalue-ai/barevalue-mcpAI podcast editing. Sign up in one call, send audio by link or from disk, get an edited episode.3stars

6

Blogburst Mcp Servershensi8312/blogburst-mcp-serverAI social media agent — generate content, auto-pilot posting, analytics for 9+ platforms.3stars