MCP Search Server

Provides 27 tools for web search, content extraction, and data processing without API keys.
  • python

3

GitHub Stars

python

Language

4 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": {
    "kazkozdev-mcp-search-server": {
      "command": "mcp-search-server",
      "args": []
    }
  }
}

You can run and explore the MCP server that provides 27 tools for web search, content extraction, and data processing without needing API keys. Its modular registry makes it easy to load, categorize, and discover tools as you work on information gathering and analysis tasks.

How to use

You interact with the MCP server through a client that can discover, query, and invoke the available tools. Start by loading the server, then use your client to search for tools by keywords, list tool categories, request detailed tool information, and run specific tools against your data. The server centralizes tool registrations, so you can dynamically find and execute the right capability for tasks like web search, content extraction, or document processing.

How to install

Prerequisites: you need Python and a working environment with network access.

# Install from PyPI (recommended)
pip install mcp-search-server

# Or install from source
git clone https://github.com/KazKozDev/mcp-search-server.git
cd mcp-search-server
pip install -e .

Additional content and configuration

Configure a client to locate and run the MCP server when you are using Claude Desktop. Create or edit the Claude configuration file to point to the MCP server command.

{
  "mcpServers": {
    "search": {
      "command": "mcp-search-server"
    }
  }
}

Notes on running locally and discovery

The MCP server uses a modular tool registry. You can discover tools, list categories, and fetch detailed tool information to understand parameters and expected outputs before running a tool.

Troubleshooting and tips

If you encounter issues starting the server, verify that Python and pip are properly installed, ensure you have network access for any required dependencies, and confirm that you are using the correct command shown in the configuration snippet.

Available tools

search_tools

Find tools by query across the registered tool registry by name, description, category, or tags and receive ranked results with relevance scores.

list_tool_categories

List all tool categories with counts to show what kinds of capabilities exist.

get_tool_info

Get detailed metadata about a specific tool, including description, parameters, category, and usage examples.

search_duckduckgo

Perform real-time web searches using DuckDuckGo with options for web or news results, plus time filtering.

search_maps

Search for places and locations, returning names, addresses, and geographic coordinates.

extract_webpage_content

Extract clean, readable text from a web page by removing ads, navigation, and boilerplate.

parse_pdf

Download and extract text from PDF documents, selecting between parsing backends automatically.

search_wikipedia

Search Wikipedia for articles matching a query and return titles, snippets, and URLs.

get_wikipedia_summary

Fetch a concise introductory summary for a Wikipedia article.

get_wikipedia_content

Retrieve the full content of a Wikipedia article in the requested language.

search_arxiv

Search arXiv for scholarly papers with optional category filtering and metadata.

search_pubmed

Search biomedical literature on PubMed with titles, abstracts, and identifiers.

search_gdelt

Query the global news database with time-based filtering for headlines and articles.

search_github

Search GitHub repositories by keywords, with sorting by popularity or update time.

get_github_readme

Fetch the README content from a GitHub repository in markdown.

search_reddit

Search Reddit posts across subreddits with optional subreddit scoping and time filters.

get_reddit_comments

Retrieve comments from a specific Reddit post, including scores and hierarchy.

get_current_datetime

Get the current date and time for a specified timezone with detailed components.

get_location_by_ip

Geolocate an IP address, returning country, city, coordinates, and ISP information.

assess_source_credibility

Evaluate credibility of a web source using multiple signals and statistical measures.

summarize_text

Summarize long text using multiple strategies with local processing and no external APIs.

calculate

Safe mathematical calculator with support for arithmetic, trigonometry, and common functions.

read_file

Read content from text, PDF, Word, Excel, or image files with automatic detection.

write_file

Write content to a file, creating it if needed, with UTF-8 text support.

append_file

Append content to an existing file or create a new one to accumulate data.

list_files

List directory contents with sizes and file types to help manage data.

delete_file

Delete a specified file within a restricted safe directory to ensure security.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational