Mastervolt Deep Research is an enterprise-grade, multi-agent research orchestration system built on VoltAgent. It coordinates a network of specialized agents to automate complex research workflows, leveraging semantic memory, custom web scraping, and financial analysis tooling to produce publication-ready reports. When you deploy it as an MCP server, the Plan Agent supervises a suite of 14+ specialized agents (such as Assistant, Writer, Scrapper, Data Analyzer, Fact Checker, Synthesizer, Coding, and Data Scientist) to perform tasks like data gathering, analysis, fact verification, and report generation. The system also includes a robust memory layer using LibSQL-backed vector storage and Google embeddings, a dedicated web scraper toolkit, and a rich UI for viewing outputs through 49 specialized components. The API layer exposes chat and messaging endpoints, enabling an interactive, streaming chat interface that feeds the agents and presents structured results.

Prerequisites:

• Node.js 18+ (and npm or pnpm/yarn)
• Git
• Google Generative AI API key (for AI capabilities)

Installation steps:

1. Clone the repository
2. Install dependencies
3. Configure environment variables

Code blocks:

Clone the repository

git clone https://github.com/ssdeanx/Mastervolt-Deep-Research.git cd Mastervolt-Deep-Research

Install dependencies

npm install

Set up environment variables

cp .env.example .env

Edit .env with your keys (example placeholders shown below)

GOOGLE_GENERATIVE_AI_API_KEY='your_google_generative_ai_api_key_here'

Optional integrations

SUPABASE_URL='your_supabase_url'

SUPABASE_KEY='your_supabase_key'

TURSO_URL='your_turso_url'

TURSO_AUTH_TOKEN='your_turso_token'

Configuration (example for local dev):

When running via MCP, you typically start through npm scripts, e.g.:

Development mode with auto-reload (Agents)

npm run dev

Start Next.js UI (Chat Interface)

npm run next

Run both concurrently

npm run dev:test

Build for production

npm run build

Start production server

npm start

Tips and common notes:

• Ensure you have a valid Google Generative AI API key added to the environment for AI capabilities.
• If you use optional integrations (Supabase, Turso), populate their URLs/keys in the environment or in the .env file.
• The MCP server expects Node.js 18+ and should be run with the provided npm scripts (dev, next, dev:test, build, start).
• When debugging, check OpenTelemetry/VoltOps traces to diagnose agent coordination issues and memory bottlenecks in the LibSQL vector store.
• If you encounter port conflicts for the Next.js UI (default 3000), adjust the PORT setting in the environment or scripts.
• Ensure the MCP server’s command path (server.js) matches your deployment layout; if your entry point differs, adjust the mcp_config accordingly.