Presearch

Presearch Search API MCP - A Model Context Protocol server for integrating Presearch search capabilities with AI assistants

javascript

GitHub Stars

javascript

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": {
    "nosytlabs-presearch-search-api-mcp": {
      "command": "npx",
      "args": [
        "presearch-mcp-server"
      ],
      "env": {
        "LOG_LEVEL": "info",
        "PRESEARCH_API_KEY": "YOUR_API_KEY",
        "PRESEARCH_TIMEOUT": "10000",
        "PRESEARCH_BASE_URL": "https://na-us-1.presearch.com"
      }
    }
  }
}

You set up a privacy‑first MCP server that bridges AI assistants to the Presearch decentralized search engine. This server lets your AI query Presearch without tracking user identities, scrape and analyze content, and run multi‑step research tasks with smart tooling.

How to use

Use the Presearch MCP Server from an MCP client to run built‑in tools like deep research, fast search and scrape, and market analysis. You can start the server locally via a runtime command, or deploy it through an MCP workflow that installs the server client‑side. Once running, configure your client to connect to the MCP endpoints and invoke the available tools to perform anonymous searches, extract content, and generate structured outputs for your AI prompts.

How to install

Install prerequisites on your machine.

Run the MCP server locally with the standard runtime command.
Deploy via Smithery if you prefer an integrated MCP client setup.

Follow the concrete commands shown below to set up and run the server in your environment.

Configuration and usage notes

## Start the server locally
npx presearch-mcp-server

## Deploy via Smithery (MCP client installation)
npx -y @smithery/cli@latest install @NosytLabs/presearch-search-api-mcp --client claude

# Optional: set up environment variables for the server
# (Replace placeholders with your actual values)
export PRESEARCH_API_KEY=YOUR_API_KEY
export PRESEARCH_BASE_URL=https://na-us-1.presearch.com
export PRESEARCH_TIMEOUT=10000
export LOG_LEVEL=info

Security and privacy considerations

The server connects to Presearch without logging user queries on disk. It uses a decentralized search network to prevent centralized IP tracking and preserves privacy in AI workflows.

Tools and capabilities overview

The server provides a suite of tools to power AI research and content extraction. You can perform standard web searches optimized for AI, autonomous multi‑step research, and targeted scraping of dynamic web content. Results can be exported in multiple formats for downstream consumption.

Available tools

presearch_ai_search

Standard web search optimized for AI queries with options like safety filters and freshness controls.

presearch_deep_research

Autonomous multi‑step research agent that performs deep investigations and topic synthesis.

presearch_search_and_scrape

Search and immediately scrape top results using a headless browser for clean content extraction.

scrape_url_content

Scrape and extract content from specified URLs with configurable text inclusion.

analyze_content

Analyze content quality and relevance, with optional keyword focus and quality metrics.

export_search_results

Export results to files in json, csv, md, html, or pdf formats.

presearch_site_export

Advanced export combining scraping and analysis for site‑level reports.

presearch_node_status

Monitor the health and status of Presearch nodes and integrations.

cache_stats

View internal cache metrics to understand performance.

cache_clear

Clear internal cache to reset state.

presearch_health_check

Verify API connectivity and readiness.

Built by

VeilStrat

AI signals for GTM teams