GreyNoise API用のMCPサーバー — IPアドレスがインターネット上のバックグラウンドノイズなのか、それとも標的型攻撃の可能性があるのかを確認します。
npx mcp-greynoiseこれだけです。APIキー不要で、1日10回のルックアップをすぐに利用できます。
Related MCP server: Yunaq GAC MCP Server
GreyNoiseは、インターネット全体のスキャントラフィックを収集・分析します。次のことを教えてくれます：
Noise：このIPはインターネットを大量スキャンしていますか？（ボットネット、研究者、クローラー）
RIOT：このIPは既知の良性サービスですか？（Google、Cloudflare、Microsoft など）
Classification：悪性、良性、または不明
ログに不審なトラフィックを確認した場合：
GreyNoiseの結果 | 解釈 |
NOISE + Malicious | バックグラウンドの攻撃トラフィック（スキャナー、ボットネット）— 標的型ではない可能性が高い |
NOISE + Benign | セキュリティ研究者、検索クローラー — 通常は安全 |
RIOT | 既知の正常なサービス（CDN、DNS、クラウド）— ほぼ確実に良性 |
NOT NOISE | ⚠️ このIPは大量スキャンではありません — トラフィックはあなたを標的にしている可能性があります |
「NOT NOISE」のケースは多くの場合最も重要です — 誰かがあなたのシステムに特に関心を持っていることを示唆しています。
check_ip の出力例：
IP: 51.91.185.74
Classification: MALICIOUS
Noise: YES - This IP has been observed scanning the internet
RIOT: NO - Not a known benign service IP
Last Seen: 2024-01-15
Details: https://viz.greynoise.io/ip/51.91.185.74
--- Interpretation ---
🚨 This IP is actively scanning the internet and classified as MALICIOUS.
Likely a scanner, botnet, or threat actor.npm install -g mcp-greynoisenpx mcp-greynoisegit clone https://github.com/nickjlucker/mcp-greynoise.git
cd mcp-greynoise
npm install
npm run build
node build/index.jsclaude_desktop_config.json に追加します：
macOS： ~/Library/Application Support/Claude/claude_desktop_config.json
Windows： %APPDATA%\Claude\claude_desktop_config.json
{
"mcpServers": {
"greynoise": {
"command": "npx",
"args": ["mcp-greynoise"],
"env": {
"GREYNOISE_API_KEY": "your-api-key-here"
}
}
}
}変数 | 必須 | 説明 |
| いいえ | より高いレート制限用のAPIキー（1日50回 vs 1日10回） |
無料のAPIキーは viz.greynoise.io/signup で取得できます。
⚠️ APIキーをコミットしないでください。 推奨される設定は
.env.exampleを参照してください。
check_ip単一のIPアドレスをGreyNoiseで確認します。
入力：
ip (string)：確認するIPv4アドレス
check_ips1回の呼び出しで複数のIPアドレスを確認します（最大10個）。
入力：
ips (string[])：IPv4アドレスの配列
出力例：
=== Results ===
8.8.8.8: RIOT (benign service) [Google]
51.91.185.74: NOISE - MALICIOUS
192.168.1.1: NOT NOISE (potentially targeted)
--- Legend ---
RIOT: Known benign service (CDN, DNS, etc.)
NOISE: IP is mass-scanning the internet
NOT NOISE: IP is NOT mass-scanning (traffic may be targeted)greynoise://statusAPIのステータスとレート制限情報を返します。
ティア | 1日のルックアップ数 |
未認証 | 10 |
無料アカウント | 50 |
有料プラン | より多い |
レート制限は、API呼び出しと GreyNoise Visualizer の間で共有されます。
このサーバーは：
GreyNoise APIから読み取りのみ行います（スキャンや悪用は行いません）
現在のリクエスト以外のデータは保存しません
APIキーをGreyNoise以外のどこにも送信しません
レピュテーション/テレメトリのエンリッチメントのみ実行します
APIキーは環境変数を介して渡され、ログに記録されることはありません。
SOCトリアージ：アラートIPがバックグラウンドノイズか標的型かを迅速に判断します
インシデントレスポンス：攻撃者IPが大量スキャナーか、焦点を絞った脅威かを特定します
脅威ハンティング：ログ内で大量スキャナーではないIP（標的型の可能性があるもの）を見つけます
ログ分析：既知のスキャナーを除外して誤検知を減らします
# Install dependencies
npm install
# Run in development mode
npm run dev
# Build for production
npm run build
# Run built version
npm startMIT
Check if an IP address is internet background noise or potentially targeted activity. GreyNoise tracks IPs that mass-scan the internet. If an IP is 'noise', traffic from it is likely untargeted scanning. If it's NOT noise, traffic may be targeted at you specifically. RIOT dataset contains known-good IPs (CDNs, DNS servers, etc.).
| Name | Required | Description | Default |
|---|---|---|---|
| ip | Yes | IPv4 address to check (e.g., 8.8.8.8) |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description bears the full burden. It explains the two classification outcomes (noise vs. targeted, and RIOT good IPs) but does not specify the actual output format, fields, or error behavior. The conceptual outline is helpful but lacks concrete details.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is concise with four sentences, front-loading the core purpose. Every sentence adds value: it defines the action, explains the background, and clarifies interpretation. No redundant or irrelevant content.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a simple lookup tool, the description covers the reasoning behind the classification, but it lacks explicit mention of the return format or fields. The existence of a sibling ('check_ips') suggests batch capability is not clarified, which could confuse an agent. The description is adequate but not fully complete.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
There is one parameter with 100% schema description coverage (example provided). The tool description adds conceptual context about what the IP is checked against, but does not add new constraints or semantic details beyond the schema. Baseline 3 is appropriate.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool checks if an IP is internet background noise or potentially targeted activity, using GreyNoise and RIOT datasets. It provides a specific verb and resource. However, it does not explicitly differentiate from the sibling tool 'check_ips', which likely handles multiple IPs.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description explains the context (security analysis, noise vs. targeted) but does not provide explicit guidance on when to use this tool over 'check_ips' or any exclusion criteria. The use case is implied but not directly stated.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Check multiple IP addresses against GreyNoise in one call. Useful for triaging a list of suspicious IPs from logs or alerts. Rate limits apply (50/day for free tier).
| Name | Required | Description | Default |
|---|---|---|---|
| ips | Yes | Array of IPv4 addresses to check (max 10) |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description carries the full burden. It discloses rate limits (50/day for free tier) and the batch nature, but does not mention idempotency, error behavior, or whether the tool is read-only. This leaves gaps in the agent's understanding of 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.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is two sentences, each earning its place: the first states the action and target, the second provides a use case and a critical constraint (rate limits). No redundancy or excess.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
The tool has 1 parameter with full schema coverage, but no output schema. The description does not explain what the tool returns (e.g., risk scores, noise classification), which is essential for an agent to interpret results. It also lacks authentication requirements or explicit differentiation from 'check_ip', leaving the agent slightly under-informed.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100%, so the baseline is 3. The description adds no new information about the 'ips' parameter beyond what the schema already provides (array of strings, max 10, min 1). It does not clarify IP format or validation rules.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the verb 'Check' and the resource 'multiple IP addresses against GreyNoise'. It also distinguishes itself from the sibling tool 'check_ip' by emphasizing batch capability ('in one call'), leaving no ambiguity about its purpose.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides clear context: 'useful for triaging a list of suspicious IPs from logs or alerts.' It also mentions rate limits. While it does not explicitly contrast with 'check_ip' or state when not to use it, the use case is well-defined and implicitly differentiates from a single-IP check.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
v0.1.0check_ipcheck_ipsScored across 2 tools
The two tools have clearly distinct purposes: one checks a single IP address, the other checks multiple IPs in bulk. An agent can easily distinguish between them based on the singular/plural naming and the explicit descriptions.
Both tools follow the consistent pattern of 'check_' followed by a noun indicating the input type ('ip' vs 'ips'). The naming is predictable and uses consistent snake_case throughout.
With only two tools, the server feels minimal for its domain. While both tools are necessary and well-scoped, a more comprehensive IP intelligence server would typically benefit from additional tools (e.g., contextual enrichment, history). The count is borderline thin.
The tool set covers the basic use case of checking IPs against GreyNoise data, both individually and in bulk. However, it lacks other common operations like fetching detailed threat context, searching by tags or queries, or retrieving known-good RIOT entries specifically. Notable gaps exist for a more complete threat intelligence surface.
GreyNoise Community MCP — internet scanner classification (free tier with key)
Free no-key IP intelligence: geolocation, VPN detection, DNS, WHOIS, blacklists, breach checks
Resolves IP autonomous system numbers (ASN) and geographic coordinates.
Enrich and search people and companies, resolve identities, and enrich IP addresses.
