sciver

MCPOfficialPipecat

Pipecat MCP Server

Gives a coding agent a voice: spoken conversation and screen viewing through Pipecat.

The Pipecat team’s server exposes voice and screen capture tools, while audio travels over a separate transport: Pipecat Playground in the browser, a Daily room or a phone line. By default it runs on local models (Faster Whisper and Kokoro) with no API keys; before talking, you start pipecat-mcp-server, which listens on localhost:9090.

Installation

Manual install

$uv tool install pipecat-ai-mcp-server && claude mcp add pipecat --transport http http://localhost:9090/mcp --scope user

Details

Works in
Claude Code, Cursor, Codex
Transport
local
License
BSD-2-Clause
Tasks
Audio, Code
Rating
Source
github.com/pipecat-ai/pipecat-mcp-server (opens in a new tab)

Tags

  • pipecat
  • voice conversation
  • voice
  • stt
  • tts
  • screen capture

Log in to rate

A pack is your own set of skills and MCP servers. sciver loads it in your AI with one command.

View in the library