M
MCP Directory CURATED DIRECTORY
911 MCP servers indexed
中文
← Back to directory
W

Web Scout MCP

Community
Search the web with DuckDuckGo and extract readable content.
How to install ↓ Compare GitHub source repository ↗
Category
Browser #21 of 32
Stars
★ 137 Very popular
Transport
stdio (local process)
Runtime
Node.js 18+
Credentials
No credential needed
License
Apache-2.0
Last commit
Tools
2
49FMRS · D

This Apache-2.0-licensed Node.js MCP server focuses on DuckDuckGo search and webpage content extraction. The README also displays an MIT badge, so the license information is inconsistent and should be verified against the repository's LICENSE file.

Strongest · Setup experience 13/20 Weakest · Reliability 8/20

Reliability
8/20
Security and permissions
9/20
Maintenance
8/20
Documentation
11/20
Setup experience
13/20
Why each score
Reliability 8/20
The README documents two tools, example inputs, the Node.js requirement, build and run commands, making the happy path plausible. However, there is no reviewable server manifest, source code, committed tests, CI, protocol-handshake evidence, or controlled error behavior. The README also shows an MIT badge while the repository metadata says Apache-2.0, reducing verifiability; the score is therefore conservative under the static rule.
Security and permissions 9/20
The stated functionality performs DuckDuckGo searches and fetches arbitrary URLs; it does not disclose credentials, writes, or destructive operations, and no real secrets appear in installation examples. However, arbitrary outbound URL fetching can create SSRF, internal-resource access, malicious-content, and data-boundary risks, while URL restrictions, network permissions, privacy handling, and a security-response channel are not documented. No red-line risk is evidenced, but the missing controls prevent a high score.
Maintenance 8/20
The project is not archived and the supplied metadata identifies an Apache-2.0 license, a repository link, and one open issue, providing a basic maintenance surface. The materials contain no release history, commit activity, issue-response evidence, dependency-update policy, security-reporting channel, or explicit maintenance commitment, so sustained maintenance is not demonstrated.
Documentation 11/20
The README covers installation methods, command-line use, mainstream client configuration, an environment variable, both tools' parameters, and basic output examples; it also links overview, contribution, and changelog documents. It omits tool limits, timeout/size/rate boundaries, privacy and cost information, authentication guidance, troubleshooting, and verifiable implementation details, and the MIT badge conflicts with the Apache-2.0 metadata.
Setup experience 13/20
The project provides npm global and local installation, an npx client configuration, and a Smithery path, states Node.js >=18, and gives concise examples for clients such as Claude Desktop and Cursor. Platform compatibility details, connection verification, troubleshooting, and test evidence are absent, and the paths depend on the external package publication state, so full marks are not justified.

Static review · not runListed 2026-07-30

Read the FMRS scoring method →

Fit and risk

What it can accessUses the network

Best for

  • AI assistant workflows that need basic web search and webpage text extraction.
  • Content-collection tasks that process multiple URLs at once.
  • MCP client users who want DuckDuckGo-based search.

Not for

  • Tasks requiring browser interaction, authenticated sessions, or page automation.
  • Workflows requiring an official search API, authentication, or a guaranteed specific search engine.
  • Tasks requiring structured webpage semantics or fully rendered page output.

Required permissions

  • Access to DuckDuckGo is needed for searches.
  • Access to user-provided target URLs is needed for content extraction.
  • A local Node.js process and stdio communication with the MCP client are required.

Risks and side effects

  • External webpage content may be incomplete because of page structure, network conditions, or site restrictions.
  • Requests may trigger rate limits or blocking from websites or search services.
  • Content retrieved from webpages should be treated as external data and verified before use.

Setup

Before you start

Runtime：Node.js 18+

Other optional settings (1)
WEB_SCOUT_DISABLE_AUTOSTART optional Set to 1 to disable automatic stdio transport startup when embedding the package and calling createServer() yourself; nothing to obtain.

Node.js 18 or later and npm or yarn are required. Install globally with npm install -g @pinkpixel/web-scout-mcp, install locally with npm install @pinkpixel/web-scout-mcp, or add the provided npx configuration to an MCP-compatible client and run web-scout-mcp. From source, run npm install, npm run build, and npm start. When embedding the package and calling createServer() manually, set WEB_SCOUT_DISABLE_AUTOSTART=1.

claude_desktop_config.json 
{"mcpServers":{"web-scout":{"command":"npx","args":["-y","@pinkpixel/web-scout-mcp@latest"]}}}

Shown for Claude Desktop. Other clients may use a different file or key (VS Code uses "servers") — the configurator below converts it.

.vscode/mcp.json
{
  "servers": {
    "web-scout": {
      "command": "npx",
      "args": [
        "-y",
        "@pinkpixel/web-scout-mcp@latest"
      ]
    }
  }
}

Goes in your project's .vscode/mcp.json (VS Code uses a "servers" key).

Terminal
claude mcp add web-scout -- npx -y @pinkpixel/web-scout-mcp@latest

Run it in a terminal; replace any <…> placeholders with your own values first.

Generate and validate client config

Check that it works

After adding the web-scout server to your client's config and restarting, DuckDuckGoWebSearch and UrlContentExtractor should appear in the tool list; running a search that returns titles, URLs, and snippets confirms the connection works.

Troubleshooting

  1. Confirm that Node.js 18 or later is installed and that npm or yarn is available.
  2. Check the MCP client configuration for the command, package name, and `-y` argument.
  3. When running from source, verify dependency installation, build, and startup steps.
  4. When embedding the package and calling `createServer()` manually, set `WEB_SCOUT_DISABLE_AUTOSTART=1`.
  5. If requests fail or are blocked, check the target URL, network connection, and rate-limit conditions.

Things to try

Once connected, you can ask your AI assistant things like:

  • Search DuckDuckGo for "latest advancements in AI" and return the top 5 results
  • Look up Model Context Protocol on DuckDuckGo and summarize the findings
  • Extract the readable content from https://example.com/article
  • Fetch the text content of https://example.com/article1 and https://example.com/article2 in parallel

Tools 2

DuckDuckGoWebSearch read-only
Runs a DuckDuckGo search and returns structured results containing titles, URLs, and snippets.
UrlContentExtractor read-only
Extracts clean, readable text from one or more webpages while removing unnecessary elements such as scripts, styles, and navigation.

Use cases

Search the web and obtain titles, links, and snippets.
Extract readable text from one or multiple webpages.
Give AI assistants web research and content-collection capabilities.

Supported clients

Claude Desktop
Cursor

Listed from the project's documentation, not tested by this site.

Overview

Web Scout MCP is a Node.js MCP server that provides DuckDuckGo web search and webpage content extraction, with support for parallel multi-URL processing, memory optimization, request throttling, and error handling.

Similar servers

Compare with other MCP servers

Source revision 4603f61ef303 Data synced 2026-10-19 Read the FMRS scoring method