voiceysurvey-mcp-server

Manas-33/voiceysurvey-mcp-server

3.2

If you are the rightful owner of voiceysurvey-mcp-server and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to dayong@mcphub.com.

The ElevenLabs Webhook MCP Server is a FastMCP server designed to handle ElevenLabs webhooks, analyze them using Gemini AI, and store the results in Firebase Firestore.

Tools
2
Resources
0
Prompts
0

ElevenLabs Webhook MCP Server

A FastMCP server that processes ElevenLabs webhooks, analyzes them with Gemini AI, and stores results in Firebase Firestore.

Setup

  1. Install dependencies:
pip install -r requirements.txt
  1. Set up environment variables in .env:
GEMINI_API_KEY=your_gemini_api_key
FIREBASE_PROJECT_ID=your_firebase_project_id
FIREBASE_PRIVATE_KEY_ID=your_private_key_id
FIREBASE_PRIVATE_KEY=your_private_key
FIREBASE_CLIENT_EMAIL=your_client_email
FIREBASE_CLIENT_ID=your_client_id
FIREBASE_AUTH_URI=https://accounts.google.com/o/oauth2/auth
FIREBASE_TOKEN_URI=https://oauth2.googleapis.com/token
  1. Run the server:
uvicorn main:app --reload --host 0.0.0.0 --port 8000

MCP server (FastMCP)

This repo also exposes your functionality via the Model Context Protocol (MCP) so IDE/chat clients can call tools directly.

  1. Install deps (already done above) and ensure mcp is installed from requirements.txt.

  2. Run over stdio (for MCP Inspector / Claude Desktop):

python mcp_server.py
  1. Or run using Streamable HTTP transport (useful for mounting into Starlette or browser clients):
MCP_TRANSPORT=http python mcp_server.py

Available MCP tools

  • process_transcript(transcript: str) → dict with request_id, analysis_id, sentiment, location.
  • generate_report(limit: int = 100) → dict shaped like the FastAPI /report response.

Demo transcript

You can use examples/demo_transcript.json as a sample payload. For FastAPI:

curl -sS -X POST http://localhost:8000/webhook/elevenlabs \
	-H "Content-Type: application/json" \
	--data @examples/demo_transcript.json

For MCP, copy the transcript string and call the process_transcript tool from your MCP client.

Endpoints

  • POST /webhook/elevenlabs - Receives ElevenLabs webhook data
  • GET /report - Generates summary report from stored data
  • GET /health - Health check endpoint

Features

  • Async webhook processing
  • Gemini AI analysis integration
  • Firebase Firestore data storage
  • Comprehensive error handling
  • Structured logging
  • Environment-based configuration