M.A.S.T.E.R

# M.A.S.T.E.R **M.A.S.T.E.R** (Mastra Academic Search Tool for Event Research) is an AI-powered orchestrator that helps you discover academic events and calls for papers based on Google Scholar author interests. The system leverages modular MCP tools and OpenAI models to automate the workflow from scholar profile to event discovery. ---  ## Features - 🔍 **Google Scholar Integration:** Fetches author interests (keywords) directly from Google Scholar using SerpAPI. - 🗝️ **Automatic Keyword Extraction:** Extracts and combines author interests into a single keyword string. - 📅 **Call-for-Papers Discovery:** Uses a custom MCP tool to find relevant academic events and calls for papers. - 🤖 **Agent-based Orchestration:** Powered by the Mastra agent framework and OpenAI models. - 🛠️ **Modular Tooling:** Easily extensible with new MCP tools. - 💬 **Telegram Integration (Coming Soon):** Planned support for Telegram bot interaction. --- ## How It Works 1. **Input:** Provide a Google Scholar `scholarId`. 2. **Interest Extraction:** The agent fetches the author's interests via SerpAPI. 3. **Event Search:** These interests are sent as keywords to the custom call-for-papers MCP tool. 4. **Output:** The agent returns a list of relevant academic events and calls for papers. --- ## Custom MCP Tool This project uses a **custom call-for-papers MCP tool** developed and published by the project author on [Smithery](https://smithery.app/). You can find and use this tool as `@alperenkocyigit/call-for-papers-mcp` on the Smithery platform. --- ## Quick Start ### 1. Clone the Repository ```bash git clone https://github.com/yourusername/M.A.S.T.E.R.git cd M.A.S.T.E.R ``` ### 2. Install Dependencies ```bash npm install ``` ### 3. Set Up Environment Variables Create a `.env` file in the root directory with the following content: ```env SMITHERY_API_KEY=your_smithery_api_key SERPAPI_API_KEY=your_serpapi_api_key SMITHERY_PROFILE_KEY=your_smithery_profile_key # (if using Telegram MCP) ``` ### 4. Run the Agent ```bash npm start ``` --- ## Usage - The agent expects a Google Scholar `scholarId` as input. - It will return a list of relevant events based on the author's interests. > **Note:** Telegram integration is planned for future releases. Currently, all features work via direct invocation. --- ## Project Structure ``` src/ mastra/ index.ts # Main agent and tool orchestration logic ... ``` --- ## MCP Tools Used - [`@alperenkocyigit/call-for-papers-mcp`](https://github.com/alperenkocyigit/call-for-papers-mcp) (custom, published on Smithery) - [`@NexusX-MCP/telegram-mcp-server`](https://smithery.app/tools/@NexusX-MCP/telegram-mcp-server) (for future Telegram integration) - SerpAPI (for Google Scholar scraping) --- ## Contributing Pull requests and suggestions are welcome! Feel free to open an issue or submit a PR. --- ## License MIT --- **Developed by [Alperen Koçyiğit](https://github.com/alperenkocyigit)**