Skip to main content
Glama
enesjakozh
PhialsBasement

MCP Web Research Server

by PhialsBasement
OverviewSchemaSourceRelated ServersDiscussionsAdmin
JavaScript
Local

MCP ウェブリサーチサーバー

Web リサーチ用のモデル コンテキスト プロトコル (MCP) サーバー。

リアルタイムの情報を Claude に取り込み、あらゆるトピックを簡単に調査できます。

特徴

  • Google 検索統合 --- このフォークでこれが修正されます --- CAPTCHA がブロックされなくなりました

  • ウェブページコンテンツの抽出

  • リサーチセッションの追跡（訪問したページのリスト、検索クエリなど）

  • スクリーンショットキャプチャ

Related MCP server: MCP Web Research Server

前提条件

インストール

まず、 Claude デスクトップ アプリをダウンロードしてインストールし、npm がインストールされていることを確認します。

次に、 claude_desktop_config.json (Mac の場合は~/Library/Application\ Support/Claude/claude_desktop_config.jsonにあります) に次のエントリを追加します。

{
  "mcpServers": {
    "webresearch": {
      "command": "npx",
      "args": ["-y", "@mzxrai/mcp-webresearch@latest"]
    }
  }
}

この設定により、Claude Desktop は必要に応じて Web リサーチ MCP サーバーを自動的に起動できるようになります。

使用法

Claudeとのチャットを開始し、Webリサーチに役立つプロンプトを送信するだけです。より深いWebリサーチ向けにカスタマイズされた、あらかじめ用意されたプロンプトが必要な場合は、このパッケージで提供されているagentic-researchプロンプトをご利用ください。Claude Desktopでこのプロンプトにアクセスするには、チャット入力欄のクリップアイコンをクリックし、 Choose an integration → webresearch → agentic-researchを選択します。

ツール

  1. search_google

    • Google検索を実行し、結果を抽出します

    • 引数: { query: string }

  2. visit_page

    • ウェブページにアクセスし、そのコンテンツを抽出する

    • 引数: { url: string, takeScreenshot?: boolean }

  3. take_screenshot

    • 現在のページのスクリーンショットを撮ります

    • 議論は必要ありません

プロンプト

agentic-research

クロードが徹底的なウェブリサーチを行うのに役立つガイド付きリサーチプロンプト。このプロンプトは、クロードに以下の指示を与えます。

  • トピックの状況を理解するために、まずは広範囲な検索から始めましょう

  • 高品質で信頼できる情報源を優先する

  • 調査結果に基づいて研究の方向性を繰り返し改善する

  • 情報を提供し、インタラクティブに研究を進めることができます

  • 常にURLでソースを引用する

リソース

MCPリソースとして、（1）キャプチャしたウェブページのスクリーンショットと（2）調査セッションの2つを公開します。

スクリーンショット

スクリーンショットを撮ると、MCPリソースとして保存されます。Claude Desktopでは、ペーパークリップアイコンからキャプチャしたスクリーンショットにアクセスできます。

研究セッション

サーバーは、次の内容を含む調査セッションを維持します。

  • 検索クエリ

  • 訪問したページ

  • 抽出されたコンテンツ

  • スクリーンショット

  • タイムスタンプ

提案

調査を行う際にagentic-researchプロンプトを使用しない場合は、クロードが一般的なトピックについて調べる際に利用できる質の高い情報源を提案すると、より効果的な結果が得られるかもしれません。例えば、 news todayではなく、 news today from reuters or APプロンプトを使うことができます。

問題

これはまだプレアルファ版のコードです。また、AIGCなのでバグが発生する可能性があります。

問題が発生した場合、Claude Desktop の MCP ログを確認すると役立つ場合があります。

tail -n 20 -f ~/Library/Logs/Claude/mcp*.log

発達

# Install dependencies
pnpm install

# Build the project
pnpm build

# Watch for changes
pnpm watch

# Run in development mode
pnpm dev

要件

  • Node.js >= 18

  • Playwright (依存関係として自動的にインストールされます)

検証済みプラットフォーム

  • [x] macOS

  • [x] リナックス

  • [x] ウィンドウズ

ライセンス

マサチューセッツ工科大学

著者

mzxrai

Available Tools

4 tools
search_googleA

Performs a web search using Google, ideal for finding current information, news, websites, and general knowledge. Use this tool when you need to research topics, find recent information, or gather data from the web. Returns structured search results with titles, URLs, and snippets.

ParametersJSON Schema
NameRequiredDescriptionDefault
queryYesSearch query

TDQS

A4/5.0
Behavior3/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

No annotations provided, so description must cover behavioral aspects. States return format (titles, URLs, snippets) but omits details like rate limits, query length limits, pagination, or error handling. Basic transparency but with notable gaps.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Three sentences, no unnecessary words, front-loaded with purpose. Every sentence adds value: what it does, when to use, what it returns.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given a simple tool with one parameter and no output schema, the description covers primary purpose, use cases, and return format. Lacks details like result count limits or error scenarios, but adequate for basic usage.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Only one parameter (query) with 100% schema description coverage (minimal: 'Search query'). The tool description adds no further parameter-specific meaning, only repeating use cases. Baseline score of 3 is appropriate.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

Clearly states 'Performs a web search using Google' with specific use cases (current information, news, websites, general knowledge). Distinguishes from sibling tools like search_scholar (academic) and visit_page (browsing).

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Provides clear guidance on when to use: 'research topics, find recent information, or gather data from the web.' Does not explicitly mention when not to use or alternatives, but context implies a general web search tool.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

search_scholarA

Searches Google Scholar for academic papers and scholarly articles. Use this tool when researching scientific topics, looking for peer-reviewed research, academic citations, or scholarly literature. Returns structured data including titles, authors, publication details, and citation counts. Ideal for academic research and evidence-based inquiries.

ParametersJSON Schema
NameRequiredDescriptionDefault
queryYesAcademic search query

TDQS

A3.7/5.0
Behavior2/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

No annotations are provided, so the description should disclose behavioral traits. It only states that it returns structured data with certain fields, but does not mention whether the tool is read-only, has rate limits, requires authentication, or any side effects. This lack of transparency is a gap.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is concise with three sentences, front-loading the core action. It is efficient and easy to parse, though slightly more structure (e.g., bullet points) could improve readability.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a simple search tool with one parameter and no output schema, the description covers purpose, usage context, and return structure. However, it lacks details on pagination, result limits, or sorting, which could be useful for an agent.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

The input schema already provides a description for the only parameter ('Academic search query'). The tool description does not add any additional meaning beyond what the schema states, so baseline 3 applies.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the tool's action ('Searches Google Scholar') and specifies the resource ('academic papers and scholarly articles'). It distinguishes itself from sibling tools like search_google by focusing on academic content.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description provides explicit guidance on when to use this tool (researching scientific topics, peer-reviewed research, etc.). It does not explicitly mention when not to use or contrast with alternatives, but the focus on academic literature implicitly differentiates it from general web searches.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

take_screenshotA

Captures a visual image of the currently loaded webpage. Use this tool when you need to preserve visual information, analyze page layouts, or document the current state of a webpage. Perfect for situations where textual content alone doesn't convey the full context.

ParametersJSON Schema
NameRequiredDescriptionDefault

No parameters

TDQS

A3.9/5.0
Behavior2/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

No annotations exist, and the description lacks details about side effects, image format, or behavior if no page is loaded, leaving significant behavioral gaps.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is two efficient sentences, front-loaded with the verb 'Captures', and every word adds value without redundancy.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness3/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

While the purpose is clear, the description omits details about output (e.g., format, full-page vs viewport), which would help completeness for a simple tool.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

With zero parameters, baseline is 4; the description correctly implies no configuration is needed.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the tool captures a visual image of the currently loaded webpage, and it is distinct from sibling tools like search_google, search_scholar, and visit_page.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description provides clear use cases (preserve visual info, analyze layouts, document state) but does not explicitly mention when not to use it or alternatives, though siblings are different.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

visit_pageA

Navigates to a specific URL and extracts the page content in readable format, with option to capture a screenshot. Use this tool to deeply analyze specific web pages, read articles, examine documentation, or verify information directly from the source. Especially useful for in-depth research after identifying relevant pages via search.

ParametersJSON Schema
NameRequiredDescriptionDefault
urlYesURL to visit
takeScreenshotNoWhether to take a screenshot

TDQS

A4.4/5.0
Behavior3/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

No annotations provided, so description must cover behavioral traits. It mentions extracting content and screenshots but omits details like error handling, dynamic content, rate limits, or side effects.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Three sentences, first covers core action and options, next two provide guidance. No unnecessary words, well-structured and front-loaded.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a simple tool with 2 params and no output schema, description adequately covers purpose and usage but lacks details on output format (e.g., how page content is returned) and potential limitations.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100% with adequate descriptions. Description adds context for takeScreenshot ('option to capture a screenshot') but does not clarify URL format or restrictions beyond the schema.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the action (navigates to URL, extracts content, optionally screenshots) and differentiates from sibling tools like search_google (which finds pages) and take_screenshot (which only captures images).

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Explicitly recommends use for in-depth research after search, and outlines scenarios (analyze pages, read articles, verify info), providing clear context for when to use.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

  1. 4 tool updates
    • First observedsearch_google
    • First observedsearch_scholar
    • First observedtake_screenshot
    • First observedvisit_page
Deploy Server
One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

TDQS

A4.2/5.0

Scored across 4 tools

Disambiguation5/5

Each tool has a clearly distinct purpose: general web search vs. academic search vs. page content extraction vs. visual capture. No overlaps in functionality.

Naming Consistency5/5

All tool names follow a consistent verb_noun pattern with snake_case (search_google, search_scholar, take_screenshot, visit_page), making them predictable.

Tool Count5/5

Four tools is well-scoped for a web research server—enough to cover core tasks without unnecessary complexity.

Completeness4/5

The set covers search (web and academic), page visiting, and screenshot capabilities. Missing a tool for managing search history or saving results, but core research workflows are supported.

Maintenance

ActivityInactive
ResponsivenessNo issues

Resources

Related MCP Connectors

Related MCP Servers

Appeared in Searches