unphurl-mcp
_ _ _ _ ____ _ _ _ _ ____ _
| | | | \ | | _ \| | | | | | | _ \| |
| | | | \| | |_) | |_| | | | | |_) | |
| |_| | |\ | __/| _ | |_| | _ <| |___
\___/|_| \_|_| |_| |_|\___/|_| \_\_____|AI 에이전트 및 개발자를 위한 URL 인텔리전스. 16개의 MCP 도구. 25개의 신호. 7가지 차원.
AI 에이전트에게 URL을 볼 수 있는 눈을 제공하세요. Unphurl은 7가지 차원(리다이렉트 동작, 브랜드 사칭, 도메인 연령, SSL/TLS, 파킹 감지, URL 구조, DNS 강화)에서 모든 URL을 분석하고, 구성 가능한 0-100점의 위험 점수와 함께 구조화된 신호를 반환합니다. 판결이 아닌 신호를 제공하며, 이를 어떻게 활용할지는 에이전트가 결정합니다.
Claude Code, Claude Desktop, Claude Cowork, ChatGPT desktop, Cursor, Windsurf 및 모든 MCP 호환 도구와 함께 작동합니다.
비용 안내
모든 신규 계정은 20개의 무료 파이프라인 확인 크레딧을 받습니다. 대부분의 조회는 무료입니다. 알려진 도메인(Tranco 상위 10만 개) 및 이전에 분석된 도메인은 비용 없이 캐시된 결과를 반환합니다. 알 수 없는 도메인이 전체 파이프라인을 통과할 때만 비용을 지불합니다.
패키지 | 크레딧 | 가격 | 확인당 비용 |
스타터 | 100 | $9 | $0.090 |
스탠다드 | 500 | $39 | $0.078 |
프로 | 2,000 | $99 | $0.050 |
스케일 | 10,000 | $399 | $0.040 |
일회성 구매이며 구독료는 없습니다. 일반적인 사용 시 URL의 95-99%는 무료로 처리됩니다.
Related MCP server: trust-score
빠른 시작
1. MCP 구성에 추가
.mcp.json 파일에 추가하세요(Claude Code, Claude Desktop, Claude Cowork, ChatGPT desktop, Cursor, Windsurf 또는 모든 MCP 호환 도구):
{
"mcpServers": {
"unphurl": {
"command": "npx",
"args": ["-y", "@unphurl/mcp-server"],
"env": {
"UNPHURL_API_KEY": "uph_your_key_here"
}
}
}
}2. 아직 계정이 없으신가요?
AI가 대신 계정을 생성할 수 있습니다. 그냥 이렇게 물어보세요: "Sign up for Unphurl." signup 도구는 API 키 없이도 작동합니다. 가입 후 MCP 구성에 키를 추가하고 재시작하세요.
3. URL 확인 시작
AI에게 말만 하세요:
"Check https://suspicious-domain.xyz"
"Batch check all URLs in this spreadsheet"
"Create a scoring profile called 'lead-qual' that weights parked domains at 30 and no MX record at 20"
"Check my credit balance"
외울 명령어는 없습니다. 배울 문법도 없습니다. AI가 도구 호출을 처리합니다.
제공되는 결과
모든 확인 결과는 7가지 차원의 신호를 반환합니다:
차원 | 설명 |
리다이렉트 동작 | 전체 체인(최대 10홉), 단축 URL 감지, 중단 사유 |
브랜드 사칭 | 150개 이상의 브랜드에 대한 Levenshtein + 호모글리프 분석 |
도메인 인텔리전스 | 연령, 등록기관, 만료일, 상태 코드, 네임서버 (RDAP 경유) |
SSL/TLS | Cloudflare 엣지에서의 실제 핸드셰이크 검증 |
파킹 감지 | 4단계 검사: 등록기관 페이지, 판매 중, 빈 콘텐츠, 파킹 서비스 |
URL 구조 | 길이, 경로 깊이, 서브도메인 수, 엔트로피, IP 감지, 인코딩된 문자 |
DNS 강화 | MX 레코드 존재 여부 (도메인이 이메일을 수신할 수 있는지 여부) |
모든 신호는 비즈니스 인텔리전스입니다. 도메인 연령은 기업의 신뢰도를 나타냅니다. MX 레코드가 없다는 것은 이메일을 수신할 수 없음을 의미합니다. 만료되는 도메인은 비즈니스가 종료될 수 있음을 시사합니다. AI의 대량 처리 및 스프레드시트 출력 기능과 결합하면 가벼운 실사 엔진이 됩니다.
16개의 도구
도구 | 기능 | 인증 |
| 새 계정 생성, API 키 획득 | 아니오 |
| 인증 이메일 재전송 (시간당 3회 제한) | 예 |
| 7가지 차원 모두에서 단일 URL 확인 | 예 |
| 최대 500개 URL 일괄 확인 (비동기 폴링 자동 처리) | 예 |
| 사용자 지정 점수 프로필 목록 확인 | 예 |
| 사용자 지정 가중치로 점수 프로필 생성 또는 업데이트 | 예 |
| 점수 프로필 삭제 | 예 |
| 기본 가중치가 적용된 25개 점수 신호 표시 | 아니오 |
| 파이프라인 확인 크레딧 잔액 확인 | 예 |
| 사용 통계 및 점수 임계값 카운트 보기 | 예 |
| 사용 가능한 크레딧 패키지 및 가격 표시 | 아니오 |
| 크레딧 구매 (Stripe 결제 URL 반환) | 예 |
| 최근 URL 확인 기록 보기 | 예 |
| 점수 증폭을 억제하는 신뢰할 수 있는 도메인 목록 확인 | 예 |
| 신뢰할 수 있는 허용 목록에 도메인 추가 (요청당 최대 100개) | 예 |
| 신뢰할 수 있는 허용 목록에서 도메인 제거 | 예 |
고급 활용법
일괄 확인 + 필터링
"Check these 500 URLs. Give me two lists: the clean ones (score under 25) and the flagged ones (score 50 or higher). Export both as CSV."
AI가 일괄 결과를 가져와 점수별로 필터링하고 목록을 출력합니다. 코드나 스크립팅은 필요 없습니다.
사용자 지정 점수 프로필
작업마다 필요한 가중치가 다릅니다. 보안 봇은 브랜드 사칭에 관심이 많고, 콜드 이메일 도구는 파킹된 도메인이나 누락된 MX 레코드에 관심이 많습니다.
"Create a profile called 'cold-email' that weights parked at 30, no_mx_record at 20, and domain_age_7 at 25. Then batch check my lead list using that profile."
다른 도구와 결합
AI 채팅 내의 Unphurl은 에이전트가 액세스할 수 있는 다른 모든 도구와 결합됩니다:
스프레드시트: CRM 내보내기 파일을 읽고, 모든 URL을 일괄 확인한 후 위험 점수와 함께 결과를 다시 기록
웹 스크래핑: 경쟁사의 파트너 페이지를 스크래핑하고 모든 링크의 상태 확인
문서: 결과로부터 브랜드화된 PDF 감사 보고서 생성
예약 작업: "매주 월요일 아침마다 내 중요 URL 확인"
결제 방식
무료 조회를 포함한 모든 확인 작업에는 양수의 크레딧 잔액이 필요합니다.
알려진 도메인(google.com, amazon.com 등): 무료, 즉시 처리.
캐시된 도메인(누군가 최근 분석한 도메인): 무료, 즉시 처리.
알 수 없는 도메인(최초 분석): 1크레딧 차감.
일괄 확인 시 알 수 없는 도메인에 대해 크레딧이 미리 차감됩니다. 크레딧이 부족하면 필요한 정확한 크레딧 양이 요약되어 표시됩니다.
실패한 파이프라인 확인은 자동으로 환불됩니다.
보조 스킬
check-url-safety 스킬은 AI가 URL을 따르거나 추천하기 전에 요청받지 않아도 사전에 확인하도록 가르칩니다.
Claude Cowork 또는 Claude Desktop: 그냥 물어보세요: *"Install the Unphurl URL safety skill."
Claude Code, Cursor 또는 기타 개발 도구:
# Global (all projects)
cp node_modules/@unphurl/mcp-server/skills/check-url-safety.md ~/.claude/skills/
# Or for a specific project
cp node_modules/@unphurl/mcp-server/skills/check-url-safety.md .claude/skills/비즈니스 구축
Unphurl은 확인당 $0.04-$0.09의 비용이 듭니다. 웹사이트 링크 감사는 15분이 소요되며 50-100 크레딧을 사용합니다. 감사당 $150-$500의 비용을 청구하세요.
서비스 아이디어: 링크 상태 감사, 리드 목록 검증, SEO 백링크 감사, 뉴스레터 링크 모니터링, 공급업체 심사 보고서, 브랜드 보호 모니터링, 인플루언서 심사.
기타 제공 형태
CLI:
npx unphurl(npm)Claude Cowork용 호스팅 MCP (로컬 설치 불필요): mcp.unphurl.com
REST API: api.unphurl.com
웹사이트: unphurl.com
시각적 가이드: unphurl.com/getting-started-guides
라이선스
MIT
Available Tools
16 toolsadd_to_allowlistAInspect
Add one or more domains to this account's trusted allowlist.
Allowlisted domains suppress the compound signal and brand impersonation floor in scoring. The full pipeline still runs — all signals remain visible so you can monitor trusted domains for SSL expiry, parking, or other changes.
Submit the registrable domain only (e.g. partnerco.com). Subdomains and full URLs are rejected. Adding partnerco.com covers sub.partnerco.com and all other subdomains automatically.
Maximum 1,000 domains per account. Maximum 100 domains per request. Duplicates are silently skipped.
Inputs: domains (required).
check_historyAInspect
View recent URL check history. Shows what URLs have been checked, their scores, phishing status, and whether each check was free or used a pipeline credit.
Results are paginated. Use page and limit parameters to navigate. Default is 20 results per page, maximum 100.
History is retained for 90 days. Account-level stats (total credits, balance) never expire.
Inputs: page, limit.
check_urlAInspect
Check a single URL for security and data quality signals. Returns a risk score (0-100), detailed signal breakdown, and metadata.
Unphurl analyses URLs across seven dimensions: redirect behaviour, brand impersonation, domain intelligence (age, registrar, expiration, status codes, nameservers via RDAP), SSL/TLS validity, parked domain detection, URL structural analysis (length, path depth, subdomain count, entropy), and DNS enrichment (MX records). The score is calculated from these signals using either default weights or a custom scoring profile.
Higher scores mean more suspicious. The score is a signal, not a verdict. You decide the threshold based on the use case.
Billing: Most lookups are free. Known domains (Tranco Top 100K like google.com, github.com) return instantly with score 0 at no cost. Previously analysed domains return cached signals at no cost. Only unknown domains that run through the full analysis pipeline cost 1 pipeline check credit. The response's meta.pipeline_check_charged field tells you whether this check consumed a credit.
Use the "profile" parameter to score results with custom weights. For example, a "cold-email" profile might weight parked domains heavily while ignoring brand impersonation. Use list_profiles to see available profiles, or show_defaults to see all signal weights.
If the account has zero credits and the URL requires a full pipeline check, returns a 402 error with a link to purchase more credits.
Inputs: url (required), profile.
check_urlsAInspect
Check multiple URLs in a single batch. Returns results for all URLs, handling async processing automatically.
Each URL is analysed across seven dimensions: redirect behaviour, brand impersonation, domain intelligence (age, registrar, expiration, status codes, nameservers via RDAP), SSL/TLS validity, parked domain detection, URL structural analysis, and DNS enrichment. Known and cached URLs return results immediately. Unknown URLs are queued for pipeline processing. This tool automatically polls for results until all URLs are complete or the 5-minute timeout is reached. You don't need to manage polling or job tracking.
If the timeout is reached before all results are complete, returns whatever is available with a clear message indicating which URLs are still processing. The user can check results later via check_history.
Maximum 500 URLs per call. For larger datasets, call this tool multiple times with chunks of up to 500 URLs.
Billing: Same as check_url. Known and cached domains are free. Only unknown domains running through the full pipeline cost 1 credit each. The summary shows pipeline_checks_charged (the actual number of credits consumed). If you don't have enough credits for the unknowns in the batch, the entire batch is rejected with a 402 error telling you exactly how many credits are needed.
Duplicate URLs in the list are automatically deduplicated (processed once, charged once). Invalid URLs get individual error status without rejecting the batch.
Use the "profile" parameter to score all results with custom weights.
Inputs: urls (required), profile.
create_profileAInspect
Create or update a custom scoring profile. Profiles are sparse overrides: only specify the weights you want to change. Everything else keeps its default value.
If a profile with this name already exists, it is updated with the new weights (full replacement, not merge).
Weights are points, not percentages. Each weight is the number of points that signal adds to the score when it fires. They don't need to total 100. A profile with weights totalling 90 is conservative (max possible score is 90). A profile with weights totalling 130 is aggressive (multiple signals quickly push to the cap of 100). The threshold the agent sets for action matters more than the weight totals.
Use show_defaults to see all 25 signals with their default weights and descriptions before creating a profile. Use check_url or check_urls with the "profile" parameter to score results with this profile.
Maximum 20 profiles per account. Profile name "default" is reserved.
Common profiles:
Cold email: weight parked (30), chain_incomplete (25), ssl_invalid (15) higher. Lower brand_impersonation (10).
Security bot: keep brand_impersonation high (40), increase domain_age_7 (30), redirects_5 (25).
Lead gen: weight parked (35), http_only (20), chain_incomplete (20) for dead business detection.
SEO audit: weight redirects_5 (30), chain_incomplete (30), parked (25) for link quality.
See the Unphurl API documentation for all 19 use case weight examples.
Inputs: name (required), weights (required).
delete_profileAInspect
Delete a custom scoring profile. This is permanent. Any future check requests using this profile name will fall back to default weights.
Use list_profiles to see your current profiles before deleting.
Inputs: name (required).
get_balanceAInspect
Check your pipeline check credit balance. Shows credits remaining, total purchased, total used, and lifetime free lookups count.
Credits are consumed only when unknown domains run through the full analysis pipeline. Known domains (Tranco Top 100K) and cached domains (previously analysed by any Unphurl customer) are always free.
If credits_remaining is 0, you can still check known and cached domains for free. To check unknown domains, purchase more credits using the "purchase" tool.
Inputs: no named parameters.
get_pricingAInspect
Show available pipeline check credit packages and pricing. Returns all packages with credit counts and prices.
Packages (one-time purchase, no subscription):
Starter: 100 credits for $9 ($0.09 each)
Standard: 500 credits for $39 ($0.078 each)
Pro: 2,000 credits for $99 ($0.0495 each)
Scale: 10,000 credits for $399 ($0.0399 each)
Most URL lookups are free (known domains and cached domains). Credits are only consumed when an unknown domain runs through the full analysis pipeline. In typical use, 95-99% of URLs resolve free.
This tool does not require an API key.
Inputs: no named parameters.
get_statsAInspect
View your account usage statistics. Shows total URLs submitted, breakdown by gate (Tranco lookups, cache lookups, pipeline checks), free rate percentage, score threshold counts, and credit balance.
Use this to understand your usage patterns: how many of your checks resolved free (known or cached domains) vs paid pipeline checks, and how many URLs scored above key thresholds.
This is useful for:
Checking if your scoring profile is flagging the right proportion of URLs
Understanding your cost efficiency (higher free rate = more value per credit)
Reporting usage metrics
Inputs: no named parameters.
list_allowlistAInspect
List all domains on this account's trusted allowlist.
Allowlisted domains suppress the compound signal and brand impersonation floor in scoring. The full pipeline still runs — all signals remain visible for monitoring. Use this to see which domains are currently trusted.
Returns the list of domains, current count, and the 1,000-domain limit.
Inputs: no named parameters.
list_profilesAInspect
List all custom scoring profiles on this account. Returns profile names and their custom weight overrides.
Profiles are named weight sets that change how Unphurl scores URLs. Different use cases need different scoring. A cold email agent cares about dead domains. A security bot cares about phishing. Profiles let one account serve multiple use cases.
Profiles only override specific weights. Any signal not specified in a profile uses the default weight. Use show_defaults to see all 25 signals and their default weights.
Inputs: no named parameters.
purchaseAInspect
Purchase pipeline check credits. Returns a Stripe Checkout URL that the user must open in a browser to complete payment.
The AI cannot complete the payment. Tell the user to open the URL in their browser, complete the Stripe checkout, and then confirm they've paid. Credits are added to the account automatically once Stripe confirms payment.
After purchase, use get_balance to verify credits have been added.
Inputs: package (required).
remove_from_allowlistAInspect
Remove one or more domains from this account's trusted allowlist.
Once removed, those domains resume normal scoring on the next check. Use list_allowlist to see what is currently on the list before removing.
Inputs: domains (required).
resend_verificationAInspect
Resend the email verification link for an existing Unphurl account.
Use this when a user signed up but their verification link expired (links are valid for 24 hours) and they need a new one. The user's API key won't work until their email is verified.
For security, the response is always the same regardless of whether the email exists, is already verified, or was rate limited. This prevents account enumeration.
Rate limited to 3 requests per email per hour.
This tool does not require an API key.
Inputs: email (required).
show_defaultsAInspect
Show all 25 scoring signals with their default weights and descriptions. This is the baseline scoring that applies when no custom profile is specified.
Use this to understand what each signal means and how much it contributes to the score before creating custom profiles. Profiles are sparse overrides on top of these defaults.
This tool does not require an API key. The defaults are hardcoded and always available.
Inputs: no named parameters.
signupAInspect
Create a new Unphurl account. Returns an API key (shown once, store it securely).
After signup, the user must check their email and click the verification link. The API key won't work for URL checks until the email is verified. Verification link expires after 24 hours. If the link expires, use the "resend_verification" tool to request a new one.
The account starts with 20 free pipeline check credits so the user can test with real URLs. Known domain lookups (google.com, github.com, etc.) and cached domain lookups are always free. To check more unknown domains through the full analysis pipeline, the user can purchase credits via the "purchase" tool.
Once the user has their API key, they need to add it to their MCP server configuration as UNPHURL_API_KEY.
This tool does not require an API key.
Inputs: email (required), first_name (required), company.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 16 tools
Tools are mostly distinct: check_url vs check_urls are clearly singular/batch, and profile management tools (list_profiles, create_profile, delete_profile, show_defaults) have clear boundaries. However, check_url/check_urls share descriptions almost verbatim verdicts (intentional but could confuse agents), and allowlist tools vs profile tools address different concepts though both modify scoring.
Tool names follow a consistent verb_noun (or imperative short) pattern: check_url, list_profiles, create_profile, get_balance, add_to_allowlist. Minor deviations: 'show_defaults' (rather than list_defaults) and 'signup' (rather than sign_up or create_account) are outliers, but overall the pattern is predictable.
With 16 tools, the count is slightly above the ideal 3-15 range but still reasonable given the breadth of functionality (auth, URL checking, billing, profiles, allowlists, history). Each tool serves a distinct purpose, and none feels redundant enough to trim, so it's borderline but acceptable.
The server covers the core lifecycle: signup/verification, URL checking (single/batch), billing (purchase, balance, pricing), profile management (CRUD plus defaults), allowlist management (CRUD), and history/stats. Missing operations include password reset or account deletion, but those are minor and not core to the URL-checking domain. The gaps are workaround-able (e.g., no explicit profile export, but not essential).
Maintenance
Related MCP Connectors
The website checker for AI agents. One call: verdict, score out of 100, fix list.
Pay-per-call safety checks for AI agents: screen a crypto address or URL before you transact.
Live web checks for AI agents: sitemaps, robots.txt, URL status, broken links, feeds, citations.
Pay-per-call APIs for AI agents: web scraping, DNS/email checks, classification, trading risk.
Related MCP Servers
- AlicenseAqualityDmaintenancesecurity tools for AI agents: URL safety scanning, prompt injection detection (200+ patterns), email/password breach checks via HIBP, domain & IP reputation analysis, and AI skill supply chain scanning. Free tier (3 calls/day) or pay-per-request with USDC micropayments via x402.919 npm1MIT
- AlicenseNot gradedqualityCmaintenanceProvides AI agents with zero-cost, zero-dependency trust evaluation for domains, URLs, wallets, APIs, and IPs, returning a 0-100 score with SSL, DNS, WHOIS, security header, and content sub-scores, plus batch comparison to rank multiple targets.MIT
- AlicenseNot gradedqualityBmaintenanceEnables AI agents to investigate public URLs through evidence-first intelligence, including crawling, entity resolution, social discovery, SEO, technology detection, monitoring, reports, and optional AI reasoning.9MIT
- AlicenseNot gradedqualityBmaintenanceEnables AI agents to inspect a URL before opening, following, or recommending it, by tracing redirect chains via headers, stripping tracking parameters, detecting phishing-style URL tricks, and cross-checking threat-intel feeds to return a SAFE, SUSPICIOUS, or DANGEROUS verdict with supporting evidence. It also scans text for embedded and defanged links and vets agent skills prior to installation.1MIT