# Research Master

> MCP to help search academic research papers via LLM

- **Type:** MCP server
- **Install:** `agentstack add mcp-hongkongkiwi-research-master`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [hongkongkiwi](https://agentstack.voostack.com/s/hongkongkiwi)
- **Installs:** 0
- **Category:** [AI & ML](https://agentstack.voostack.com/c/ai-and-ml)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [hongkongkiwi](https://github.com/hongkongkiwi)
- **Source:** https://github.com/hongkongkiwi/research-master

## Install

```sh
agentstack add mcp-hongkongkiwi-research-master
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# Research Master

A Model Context Protocol (MCP) server for searching and downloading academic papers from multiple research sources.

[](https://opensource.org/licenses/MIT)
[](https://www.rust-lang.org/)
[](https://crates.io/crates/research-master)
[](https://github.com/hongkongkiwi/research-master/releases)

## Overview

Research Master is a comprehensive academic research server that provides unified access to **28 major research repositories and databases**. It implements the Model Context Protocol (MCP) to integrate seamlessly with AI assistants like Claude Desktop, enabling powerful literature search, paper discovery, and citation analysis capabilities.

## Quick Start

### 1. Install

**macOS (Homebrew):**
```bash
brew tap hongkongkiwi/research-master
brew install research-master
```

**Other methods:** See [Installation](docs/installation.md) for Linux packages, Docker, and building from source.

### 2. Configure Your MCP Client

Add to your Claude Desktop configuration (`~/Library/Application Support/Claude/claude_desktop_config.json`):

```json
{
  "mcpServers": {
    "research-master": {
      "command": "research-master",
      "args": ["mcp"]
    }
  }
}
```

See [MCP Clients](docs/mcp-clients.md) for setup instructions for Claude Desktop, Zed, Cursor, Continue, and 15+ other clients.

### 3. Start Searching

Once configured, use natural language with your AI assistant:

```
Search for papers about "transformer architecture" from 2020 onwards
Download the paper 1706.03762 and find what papers cite it
Find papers by Geoffrey Hinton on deep learning
```

See [Tools](docs/tools.md) for all available MCP tools.

## Features

- **28 Research Sources**: arXiv, Semantic Scholar, OpenAlex, PubMed, PMC, bioRxiv, and [more](docs/sources.md)
- **Unified Search**: Single query searches across all sources
- **Smart Source Detection**: Automatically identifies paper IDs (arXiv, PMC, DOI, etc.)
- **PDF Download**: Save papers to your local filesystem
- **Citation Analysis**: Find papers that cite or are cited by a paper
- **Deduplication**: Remove duplicate results across sources
- **Rate Limiting**: Configurable to avoid API throttling

See [Sources](docs/sources.md) for supported databases, API requirements, and rate limits.

## Documentation

| Topic | Description |
|-------|-------------|
| [Installation](docs/installation.md) | Install via Homebrew, Docker, packages, or source |
| [Sources](docs/sources.md) | Supported research databases, API keys, rate limits |
| [Usage](docs/usage.md) | CLI commands and options |
| [Tools](docs/tools.md) | Available MCP tools reference |
| [MCP Clients](docs/mcp-clients.md) | Configuration for Claude Desktop, Zed, Cursor, etc. |
| [Configuration](docs/configuration.md) | Environment variables and config file |
| [Development](docs/development.md) | Project structure, adding new sources |

## Common Commands

```bash
# Search for papers
research-master search "transformer architecture" --year 2020-

# Search by author
research-master author "Geoffrey Hinton"

# Download a paper
research-master download 2301.12345 --source arxiv --output ./papers

# Look up by DOI
research-master lookup 10.48550/arXiv.2301.12345

# Start MCP server
research-master mcp --stdio

# Show all environment variables
research-master --env
```

## Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

### Ways to Contribute

- Report bugs and issues
- Suggest new features
- Add new research sources
- Improve documentation
- Submit pull requests

See [Development](docs/development.md) for the project structure and how to add new sources.

## License

This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details.

## Acknowledgments

- Built with [Rust](https://www.rust-lang.org/)
- Implements the [Model Context Protocol](https://modelcontextprotocol.io/)
- Integrates with numerous academic research APIs and services

## Contact

- GitHub: [@hongkongkiwi](https://github.com/hongkongkiwi)
- Repository: [https://github.com/hongkongkiwi/research-master](https://github.com/hongkongkiwi/research-master)

## Source & license

This open-source MCP server is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [hongkongkiwi](https://github.com/hongkongkiwi)
- **Source:** [hongkongkiwi/research-master](https://github.com/hongkongkiwi/research-master)
- **License:** MIT

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/mcp-hongkongkiwi-research-master
- Seller: https://agentstack.voostack.com/s/hongkongkiwi
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.