Keyword Search MCP Server

# ✅ MCP Server – Keyword Search Tool (Case-Insensitive) This project is a simple **MCP (Model Context Protocol) Server** built using FastAPI. It includes a tool that allows users to **search for a keyword inside a file** (case-insensitive). --- ## 📄 Documentation You can view the detailed documentation here: **[Assignment Documentation](https://docs.google.com/document/d/1prMqMwjT5Zi70FqnQzCr1r0WSaW4CHpoGg6EjazEonM/edit?usp=sharing)** --- ## 📁 Project Structure ``` mcp-server/ ├── server.py ├── tools/ │ ├── __init__.py │ └── search_tool.py ├── requirements.txt └── README.md ``` --- ## ⚙️ 1. Setup Instructions ### ✅ Step 1: Clone or Create Project Folder ```bash cd mcp-server ``` ### ✅ Step 2: Create and Activate Virtual Environment (Optional but Recommended) ```bash python -m venv venv # Activate it: # On Windows: venv\Scripts\activate # On Linux/Mac: source venv/bin/activate ``` ### ✅ Step 3: Install Dependencies ```bash pip install -r requirements.txt ``` --- ## 🚀 2. Run the MCP Server ```bash uvicorn server:app --reload --port 8000 ``` If successful, you will see: ``` Uvicorn running on http://127.0.0.1:8000 ``` Visit in browser: ``` http://127.0.0.1:8000 ``` Or API docs (auto-generated): ``` http://127.0.0.1:8000/docs ``` --- ## 🔍 3. Search Keyword in a File (API Usage) ### ✅ Request Format (POST `/search`) ```json { "file_path": "path/to/file.txt", "keyword": "example" } ``` ### ✅ Example using `curl`: ```bash curl -X POST "http://127.0.0.1:8000/search" \ -H "Content-Type: application/json" \ -d '{"file_path": "sample.txt", "keyword": "hello"}' ``` ### ✅ Example Response: ```json { "results": [ "Line 2: Hello World", "Line 5: hElLo again" ] } ``` --- ## 🛠 4. Code Summary ### `server.py` Defines API endpoint `/search`. --- ## ✅ 5. Features ✔ FastAPI-powered MCP server ✔ Case-insensitive keyword search ✔ Shows line number + text where keyword exists ✔ Error handling for missing files --- ## 📌 6. Future Improvements (Optional) * 🔹 Search across directories * 🔹 Regex support * 🔹 Return JSON with `{line_number, content}` format * 🔹 File type filtering (.txt, .py, .md, etc.)