Énumérez les enregistrements DNS, auditez les protocoles de sécurité e-mail et détectez les vulnérabilités de détournement de sous-domaine.
Aussi connu sous GuardBee/guardbee-mcp, @guardbee/mcp-dns-intelligence
- Étoiles
- 2
- Outils
- 6
- Téléchargements, 30 jours
- 2 296
- Dernière version
- 0.2.3
- Dernier commit
- 12 nov. 2026
- Licence
- MIT
Qu'est-ce que dns-intelligence MCP ?
Énumérez les enregistrements DNS, auditez les protocoles de sécurité e-mail et détectez les vulnérabilités de détournement de sous-domaine. Il expose 6 outils, comme enumerate_dns, audit_domains, enumerate_subdomains.
Résumé généré à partir du README du projet
À propos du MCP server dns-intelligence
Les chercheurs en sécurité et les développeurs peuvent effectuer des audits DNS approfondis et des évaluations de la sécurité des e-mails directement via un client IA. Ce serveur fournit des outils pour énumérer divers enregistrements DNS, notamment A, MX et TXT, tout en attribuant une note de A à F basée sur la posture de sécurité du domaine. Il cible spécifiquement l'authentification des e-mails en analysant les configurations SPF, DMARC et DKIM afin d'identifier les erreurs de configuration, comme des recherches excessives ou l'absence de politiques.
Au-delà de la recherche d'enregistrements standard, le serveur identifie les vulnérabilités potentielles de détournement de sous-domaine en scannant les journaux Certificate Transparency et en recherchant des enregistrements CNAME orphelins. Il peut sonder les sélecteurs DKIM courants et évaluer l'intégrité des paramètres DNSSEC et CAA. Les utilisateurs peuvent exécuter ces vérifications localement via npx ou les intégrer dans leurs flux de travail à l'aide des outils MCP fournis. Le système est conçu pour être léger, utilisant les modules intégrés de Node.js sans dépendances externes, et maintient la télémétrie désactivée par défaut.
Ce que vous pouvez faire avec
- 1Auditer les protocoles de sécurité e-mailAnalysez les paramètres SPF, DMARC et DKIM d'un domaine pour recevoir des recommandations de correction spécifiques.
- 2Détecter les risques de détournement de sous-domaineIdentifiez les sous-domaines vulnérables à un détournement en recherchant des enregistrements CNAME orphelins.
- 3Effectuer une énumération DNS complèteObtenez un rapport complet de tous les enregistrements DNS et une note de sécurité pour un domaine spécifique.
- 4Comparer les notes de sécurité de plusieurs domainesÉvaluez jusqu'à 20 domaines différents simultanément pour comparer leurs résultats de sécurité.
Extrait du README du projet. Chaque cas d'utilisation est vérifié par rapport à une citation du document.
Comment installer dns-intelligence MCP
Formats vérifiés le 10 nov. 2026Ajouter à CursorAjouter à GooseInstaller dns-intelligence MCP dans Claude Code
$ claude mcp add dns-intelligence -- npx -y @guardbee/mcp-dns-intelligenceInstaller dns-intelligence MCP dans Cursor
~/.cursor/mcp.json
{
"mcpServers": {
"dns-intelligence": {
"command": "npx",
"args": [
"-y",
"@guardbee/mcp-dns-intelligence"
]
}
}
}Installer dns-intelligence MCP dans VS Code
.vscode/mcp.json
{
"servers": {
"dns-intelligence": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"@guardbee/mcp-dns-intelligence"
]
}
}
}Installer dns-intelligence MCP dans VS Code CLI
$ code --add-mcp '{"name":"dns-intelligence","type":"stdio","command":"npx","args":["-y","@guardbee/mcp-dns-intelligence"]}'Installer dns-intelligence MCP dans Claude Desktop
claude_desktop_config.json
{
"mcpServers": {
"dns-intelligence": {
"command": "npx",
"args": [
"-y",
"@guardbee/mcp-dns-intelligence"
]
}
}
}Installer dns-intelligence MCP dans Codex
$ codex mcp add dns-intelligence -- npx -y @guardbee/mcp-dns-intelligenceInstaller dns-intelligence MCP dans Codex config.toml
~/.codex/config.toml
[mcp_servers.dns-intelligence] command = "npx" args = ["-y", "@guardbee/mcp-dns-intelligence"]
Installer dns-intelligence MCP dans Gemini CLI
$ gemini mcp add -s user dns-intelligence npx -- -y @guardbee/mcp-dns-intelligenceInstaller dns-intelligence MCP dans Gemini settings.json
~/.gemini/settings.json
{
"mcpServers": {
"dns-intelligence": {
"command": "npx",
"args": [
"-y",
"@guardbee/mcp-dns-intelligence"
]
}
}
}Outils MCP dns-intelligence 6
Détectés depuis le READMEenumerate_dnsaudit_domainsenumerate_subdomainscheck_takeovercheck_email_securitylookup_dns
| enumerate_dns | — | |
| audit_domains | — | |
| enumerate_subdomains | — | |
| check_takeover | — | |
| check_email_security | — | |
| lookup_dns | — |
Vérifications automatisées
Exécuté le 13 nov. 2026Rapport complet →A passé nos vérifications automatisées. Il ne s'agit pas d'un audit de sécurité : lisez le code et donnez au serveur l'accès le plus restreint dont il a besoin. Fonctionnement des vérifications
- Dépôt source lié et accessibleRéussi
- Licence open source (OSI)Réussi
- Commit ou version publiée au cours de l'année écouléeRéussi
- Nom du package non proche d'un package populaireRéussi
- L'endpoint distant utilise HTTPSNon vérifié
- Les outils destructifs sont étiquetésRéussi
- Aucune instruction cachée destinée au modèleRéussi
- Aucun avis critique dans le package ou ses dépendancesRéussi
README
Voir sur GitHub@guardbee/mcp-dns-intelligence
🇬🇧 English | 🇹🇷 Türkçe | 🇨🇳 中文 | 🇯🇵 日本語 | 🇰🇷 한국어 | 🇧🇷 Português | 🇪🇸 Español | 🇩🇪 Deutsch
An MCP server that enumerates DNS records, detects SPF / DMARC / DKIM misconfigurations, and finds subdomains (Certificate Transparency + wordlist) that can be taken over. Uses Node.js's built-in dns/promises — no external dependencies.
Telemetry is off by default. With
GUARDBEE_TELEMETRY=1it sends the server, tool, success, duration and short parameter values such as a file path; longer strings and fields likecontent,tokenorpasswordare replaced with a placeholder. See@guardbee/mcp-telemetry.
Features
- Full DNS Enumeration — A, AAAA, MX, NS, TXT, CNAME, SOA records
- SPF Analysis —
+all,?all, duplicate records, and the RFC 7208 limits counted the way receivers do: DNS lookups across nestedinclude/redirect(max 10), void lookups (max 2), include loops and missing includes - DMARC Analysis — policy (
none/quarantine/reject),pct, missingruaaddress - DKIM Check — probes 9 common selectors (
default,google,selector1,mail, etc.) - Subdomain Discovery — names from Certificate Transparency logs (crt.sh and Cert Spotter, queried in parallel; only the domain name is sent) plus a 75-name wordlist, with wildcard DNS filtered out; every live name is checked for subdomain takeover
- Email Security Summary — combined SPF + DMARC + DKIM analysis with copy-pasteable fix recommendations
- 23 Unit Tests — pure logic tests, no network connection required
Quick Start
npm install -g @guardbee/mcp-dns-intelligence
Add to claude_desktop_config.json:
{
"mcpServers": {
"guardbee-dns-intelligence": {
"command": "npx",
"args": ["-y", "@guardbee/mcp-dns-intelligence"]
}
}
}
MCP Tools
| Tool | Description |
|---|---|
enumerate_dns |
Full DNS check with an A–F grade: records, SPF/DMARC/DKIM, MTA-STS, DNSSEC, CAA, nameservers |
audit_domains |
Audits up to 20 domains and compares their grades and top findings |
enumerate_subdomains |
Probes common subdomains (wildcard-aware); detects subdomain takeovers |
check_takeover |
Checks given hostnames for subdomain takeover |
check_email_security |
Combined SPF + DMARC + DKIM audit with fix recommendations |
lookup_dns |
Targeted DNS lookup for a specific record type (A/MX/TXT/etc.) |
Example Usage
You can ask Claude:
"Is there a problem with example.com's DNS configuration?"
"Audit example.com's email security — SPF, DMARC, and DKIM"
"List example.com's subdomains, flag any dangling ones"
"What are example.com's MX records?"
Example Output
Email Security Check: example.com
──────────────────────────────────────────────────
── SPF ─────────────────────────────────────────
✓ v=spf1 include:_spf.google.com -all
── DMARC ───────────────────────────────────────
✗ No DMARC record found at _dmarc.example.com
Add: v=DMARC1; p=quarantine; rua=mailto:dmarc@example.com
── DKIM ────────────────────────────────────────
✓ Found DKIM selectors: google
SPF Findings
| Code | Severity | Description |
|---|---|---|
SPF_MISSING |
🟡 Medium | No SPF record found |
SPF_PERMISSIVE_ALL |
🔴 Critical | +all — allows any server |
SPF_NEUTRAL_ALL |
🟠 High | ?all — doesn't reject unauthorized senders |
SPF_NO_ALL |
🟡 Medium | No all mechanism |
SPF_TOO_MANY_LOOKUPS |
🟠 High | > 10 DNS lookups — causes SPF to fail |
SPF_DUPLICATE |
🟠 High | Multiple SPF records |
SPF_AT_LOOKUP_LIMIT |
🔵 Low | Exactly 10 lookups — one more include breaks SPF |
SPF_TOO_MANY_VOID_LOOKUPS |
🟠 High | More than 2 includes return no record |
SPF_INCLUDE_MISSING |
🟠 High | An include/redirect target has no SPF record (permerror) |
SPF_INCLUDE_DUPLICATE |
🟠 High | An included domain publishes several SPF records |
SPF_INCLUDE_LOOP |
🟠 High | Includes loop back to a record being evaluated |
SPF_PTR |
🔵 Low | Uses the deprecated ptr mechanism |
DMARC Findings
| Code | Severity | Description |
|---|---|---|
DMARC_MISSING |
🟠 High | No DMARC record |
DMARC_NO_POLICY |
🟠 High | Missing p= policy |
DMARC_POLICY_NONE |
🟡 Medium | p=none — monitoring mode, no enforcement |
DMARC_PCT_LOW |
🔵 Low | pct < 100 — partial enforcement |
DMARC_NO_RUA |
🔵 Low | No aggregate report address (rua) |
DMARC_MULTIPLE |
🟠 High | Several DMARC records — receivers ignore all of them |
DMARC_INVALID_POLICY |
🟠 High | p= is not none, quarantine or reject |
DMARC_SUBDOMAINS_NONE |
🟡 Medium | sp=none leaves subdomains spoofable |
DOMAIN_NOT_FOUND |
🟠 High | The domain does not exist (NXDOMAIN) |
Subdomains without their own DMARC record inherit the parent domain's (sp= policy), as receivers do. Subdomain enumeration detects wildcard DNS and leaves out names answered only by the wildcard.
DKIM, MX and Transport Findings
| Code | Severity | Description |
|---|---|---|
DKIM_NOT_FOUND |
🟡 Medium | No DKIM key under ~35 common selectors (pass yours via dkimSelectors) |
DKIM_KEY_WEAK |
🟠 High | RSA key under 1024 bits |
DKIM_KEY_1024 |
🟡 Medium | 1024-bit RSA key — rotate to 2048 (RFC 8301) |
DKIM_KEY_INVALID |
🟡 Medium | Published key cannot be parsed |
DKIM_TESTING |
🔵 Low | t=y test mode |
MX_UNRESOLVABLE |
🟠 High | An MX host has no A/AAAA record |
MTA_STS_MISSING |
🔵 Low | No MTA-STS: inbound mail can be downgraded to plaintext |
MTA_STS_MX_MISMATCH |
🟠 High | Enforce policy omits a live MX — compliant senders refuse to deliver |
MTA_STS_POLICY_UNREACHABLE / _INVALID |
🟡 Medium | Policy file missing, redirected or malformed — senders ignore MTA-STS |
MTA_STS_TESTING / _MODE_NONE / _MAX_AGE_SHORT |
🔵 Low | Not enforcing, withdrawn, or max_age under a day |
TLSRPT_MISSING / _INVALID |
🔵 Low | No usable TLS-RPT next to MTA-STS |
SPF_MISSING is 🟠 High when the domain has MX records (with a starting record, e.g. v=spf1 include:_spf.google.com ~all for Google Workspace) and 🟡 Medium otherwise (suggests v=spf1 -all).
Nameserver Findings
Each nameserver of a zone is queried directly over DNS (UDP/TCP 53): an SOA query without recursion, an AXFR request (only the first messages are read — the zone is never downloaded), and a recursive query for a name it does not host. Skip with --no-ns-probe / nameserverProbe: false.
| Code | Severity | Description |
|---|---|---|
NS_AXFR_ALLOWED |
🟠 High | Anyone can download the whole zone (zone transfer) |
NS_LAME |
🟠 High | A listed nameserver does not answer authoritatively (REFUSED, SERVFAIL, no AA, or no answer) |
NS_UNRESOLVABLE |
🟠 High | A nameserver's own name has no address |
NS_OPEN_RESOLVER |
🟡 Medium | The nameserver answers recursive queries for anyone (amplification, cache poisoning) |
NS_SOA_SERIAL_MISMATCH |
🔵 Low | Servers of the same primary return different serials (multi-provider setups are compared per primary) |
DNSSEC and CAA Findings
| Code | Severity | Description |
|---|---|---|
DNSSEC_BROKEN |
🔴 Critical | Validation fails — validating resolvers (Google, Cloudflare, most ISPs) cannot resolve the domain |
DNSSEC_NO_DS |
🟡 Medium | Zone is signed but the parent has no DS, so nothing is validated |
DNSSEC_WEAK_ALGORITHM |
🟡 Medium | Deprecated signing algorithm (RSASHA1, RSAMD5, DSA, GOST; RFC 8624) |
DNSSEC_DS_SHA1 |
🔵 Low | The only DS digest is SHA-1 |
DNSSEC_DISABLED |
🔵 Low | Zone is not signed |
CAA_BLOCKS_CURRENT_CA |
🟡 Medium | CAA does not allow the CA that issued the live certificate — its renewal will fail |
CAA_UNKNOWN_CRITICAL |
🟠 High | A critical CAA tag CAs do not understand — every CA must refuse |
CAA_MISSING |
🔵 Low | No CAA anywhere up the tree: any CA may issue |
DNSSEC is read through DNS-over-HTTPS (dns.google), comparing a validating query with a cd=1 (checking disabled) one, because Node's resolver cannot see DS/DNSKEY records or the AD flag. CAA follows RFC 8659 (nearest ancestor), and the issuer of the certificate served on :443 is compared against it. Both checks are skipped with http: false.
Subdomain Takeover Detection
Each live subdomain with a CNAME is checked, following the CNAME chain (up to 5 hops). The wildcard record is checked too: a dangling *.domain makes every unlisted name claimable.
| Risk | When | Providers |
|---|---|---|
| ⚠️ takeover possible | The provider lets anyone claim the name | AWS S3 (NoSuchBucket page), Microsoft Azure and AWS Elastic Beanstalk (CNAME target NXDOMAIN), Surge.sh (project not found), and any CNAME into a domain that is not registered |
| ⚠️ verify manually | The resource is gone, but the provider can require domain verification | GitHub Pages, Heroku, Netlify, Vercel, Ghost, Pantheon, WordPress.com, Help Scout, Tumblr, ngrok |
| 🧹 stale | The target is gone but sits under a registered domain only its owner controls — delete the record; not a takeover, does not fail CI | any other provider |
Each page fingerprint was checked against a random unclaimed name on the provider. Providers whose "not found" page no longer carries a stable marker are not listed. HTTP checks send one GET / to the subdomain (HTTPS, then HTTP); turn them off with --no-http / httpProbe: false. Use check_takeover / takeover <host…> for hostnames you already know.
CLI — CI/CD Integration
In addition to MCP server mode, this can also be used directly as a CLI:
# Full DNS + email security check
npx @guardbee/mcp-dns-intelligence check example.com
# Fail only on high and above
npx @guardbee/mcp-dns-intelligence check example.com --fail-on=high
# Subdomain scan — exits 1 if a dangling CNAME is found
npx @guardbee/mcp-dns-intelligence subdomains example.com
# Subdomain scan with higher concurrency
npx @guardbee/mcp-dns-intelligence subdomains example.com --concurrency=50
# JSON output
npx @guardbee/mcp-dns-intelligence check example.com --format=json
# Several domains with an A–F grade each and a summary table (or every domain in guardbee.yml when none is given)
npx @guardbee/mcp-dns-intelligence check example.com example.de example.co.uk
npx @guardbee/mcp-dns-intelligence check --format=sarif > dns.sarif
Grades: each finding deducts points (critical 40, high 20, medium 8, low 2) and the worst finding caps the grade — any critical is F, any high at most C, any medium at most B.
Exit codes: 0 = no issues · 1 = findings above threshold / dangling subdomain · 2 = error
Configuration via guardbee.yml
Create a guardbee.yml at your project root to persist CLI flags. CLI flags always override file settings.
dns-intelligence:
fail-on: high # critical | high | medium | low
concurrency: 20 # parallel subdomain probes
domains: # used when no domain is given on the CLI
- example.com
- staging.example.com
See guardbee.example.yml for a sample file.
GitHub Actions — DNS Security Audit
name: DNS Security Check
on:
schedule:
- cron: "0 6 * * *" # Every day at 06:00
workflow_dispatch:
jobs:
dns-check:
runs-on: ubuntu-latest
steps:
- name: Check DNS configuration
run: npx @guardbee/mcp-dns-intelligence check ${{ vars.DOMAIN }} --fail-on=high
- name: Scan for dangling subdomains
run: npx @guardbee/mcp-dns-intelligence subdomains ${{ vars.DOMAIN }}
GitLab CI
dns-security:
image: node:20
script:
- npx @guardbee/mcp-dns-intelligence check $DOMAIN --fail-on=high
- npx @guardbee/mcp-dns-intelligence subdomains $DOMAIN
only:
- schedules
Pre-Deployment Email Security Check
- name: Verify email security records
run: |
npx @guardbee/mcp-dns-intelligence check ${{ vars.DOMAIN }} \
--fail-on=high \
--format=json | tee dns-report.json
- name: Upload DNS report
uses: actions/upload-artifact@v4
with:
name: dns-security-report
path: dns-report.json
Development
npm install
npm test # 23 unit tests
npm run build # TypeScript compile
License
MIT — GuardBee
Feedback
Questions, ideas, or a false positive? Start a thread in GitHub Discussions or open an issue. If GuardBee is useful to you, a ⭐ on the repository helps others find it.
Questions sur dns-intelligence MCP
Comment ajouter dns-intelligence MCP à Claude Code ?
Exécutez claude mcp add dns-intelligence -- npx -y @guardbee/mcp-dns-intelligence dans votre terminal. Ajoutez --scope user pour le rendre disponible dans tous les projets.
Comment ajouter dns-intelligence MCP à Cursor ?
Ouvrez ~/.cursor/mcp.json et ajoutez le serveur sous mcpServers, avec la configuration Cursor indiquée dans la section d'installation ci-dessus.
Quels outils dns-intelligence MCP fournit-il ?
Il expose 6 outils, dont enumerate_dns, audit_domains, enumerate_subdomains, check_takeover, check_email_security.
Quels enregistrements DNS peuvent être énumérés ?
Le serveur prend en charge l'énumération complète des enregistrements A, AAAA, MX, NS, TXT, CNAME et SOA.
Comment l'outil détecte-t-il les détournements de sous-domaine ?
Il sonde les sous-domaines courants en utilisant les journaux Certificate Transparency et une liste de mots, puis vérifie les cibles CNAME actives pour identifier les ressources orphelines pouvant être revendiquées.
Le serveur envoie-t-il des données de télémétrie ?
La télémétrie est désactivée par défaut ; elle peut être activée en utilisant la variable d'environnement GUARDBEE_TELEMETRY=1.
Puis-je auditer spécifiquement la sécurité des e-mails ?
Oui, le serveur propose un outil pour effectuer un audit combiné SPF, DMARC et DKIM avec des recommandations de correction.