MCPower

Knowledge Server

typescript

1

GitHub Stars

typescript

Language

6 months ago

First Indexed

2 months ago

Catalog Refreshed

Documentation & install

Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.

Installation

Add the following to your MCP client configuration file.

Configuration

View docs

{
  "mcpServers": {
    "wspotter-mcpower": {
      "command": "node",
      "args": [
        "/absolute/path/to/mcpower/dist/cli.js",
        "--datasets",
        "./datasets"
      ],
      "env": {
        "MCPOWER_PYTHON": "/absolute/path/to/mcpower/.venv/bin/python"
      }
    }
  }
}

MCPower is a Model Context Protocol (MCP) server that lets you search and explore your document collections with semantic understanding. By indexing your TXT and MD files into FAISS-based vector embeddings, you can query your knowledge bases through a responsive web UI or programmatic MCP tools, enabling fast, natural-language search across multiple datasets.

How to use

You connect to MCPower from any MCP client to perform knowledge searches against your datasets. Use the built-in tools to search across one or more knowledge bases, list available datasets, or fetch contextual results with relevance scores. You can manage datasets directly from the web console or via the MCP tools exposed by MCPower. The system is designed to degrade gracefully if a dataset is invalid or unavailable, so you can experiment with minimal setup and add more sources over time.

How to install

Prerequisites: you need Node.js 18+ and npm, a Python 3.10+ installation, and Git.

Additional notes

Configuration is handled through command-line options and environment variables. The typical flow is to index your documents into datasets, run the development server, and connect MCP clients to the provided MCP endpoints. You can start the web console, monitor indexing progress in real time, and use the MCP tools to perform semantic searches.

Available tools

knowledge.search

Search your knowledge bases using natural language queries and return ranked results with optional topK.

knowledge.listDatasets

List all available datasets with basic metadata and statistics.