MCP HubMCP Hub
SKILL·D0AD32

email-deliverability

rampstackco
업데이트됨 25 days ago
6 조회
953
132
953
GitHub에서 보기
메타ai

정보

이 스킬은 인증 프로토콜(SPF, DKIM, DMARC) 구현과 발신자 평판 모니터링을 통해 개발자들이 이메일 전달성을 보장하도록 돕습니다. 스팸 배치 문제 진단, 스푸핑 방지 강화, 전달성 감사 수행을 지원합니다. 트랜잭션이나 마케팅 이메일이 사용자에게 도달하지 않거나 새 도메인에 대한 이메일을 구성할 때 사용하세요.

빠른 설치

Claude Code

추천
기본
npx skills add rampstackco/claude-skills -a claude-code
플러그인 명령대체
/plugin add https://github.com/rampstackco/claude-skills
Git 클론대체
git clone https://github.com/rampstackco/claude-skills.git ~/.claude/skills/email-deliverability

Claude Code에서 이 명령을 복사하여 붙여넣어 스킬을 설치하세요

문서

Email Deliverability

Get email into inboxes, not spam folders. Set up authentication. Monitor reputation. Diagnose problems before they hurt the business.


When to use

  • Setting up email for a new domain
  • A meaningful percentage of email is going to spam
  • Customers report they're not receiving emails
  • Setting up DMARC, SPF, or DKIM
  • Hardening against domain spoofing
  • Migrating email service providers
  • Sender reputation has dropped
  • Pre-launch audit before sending volume increases

When NOT to use

  • Writing the email content itself (use email-sequences)
  • Designing the email program strategy (use email-sequences)
  • DNS records in general (use domain-strategy)
  • Outbound spam coming FROM your account (different problem; investigate compromised credentials)

Required inputs

  • The sending domain(s)
  • The email service provider (ESP, transactional service, mail server)
  • Current DNS records (or access to them)
  • Email volume (transactional vs marketing, daily volume)
  • Current deliverability state (if known: bounce rate, spam complaints)

The framework: 3 pillars

Email deliverability rests on three pillars. Weakness in any one limits the others.

Pillar 1: Authentication

Mailbox providers verify email is actually from who it claims to be from. Three records.

SPF (Sender Policy Framework)

Lists which servers are authorized to send mail for the domain. Published as a TXT record at the apex.

v=spf1 include:_spf.mailprovider.com -all
  • include: adds another sender's authorized list
  • -all (hard fail): mail from unlisted senders fails authentication
  • ~all (soft fail): unlisted senders are suspicious but pass; useful during rollout
  • +all: never use; allows anyone to send

Only one SPF record per domain. Multiple SPF records break SPF entirely. Combine senders into a single record.

SPF has a 10-DNS-lookup limit. Each include: may use multiple lookups. Hit the limit and SPF stops working. Watch this carefully.

DKIM (DomainKeys Identified Mail)

A cryptographic signature on each outgoing email. The mail server signs with a private key; the public key is published in DNS.

selector1._domainkey.example.com    TXT    "v=DKIM1; k=rsa; p=MIGfMA0G..."

Selectors differ by ESP. Some use default._domainkey, some use unique selectors per service. Most ESPs walk you through publishing the records.

DKIM proves the message wasn't modified in transit and that the sender controls the domain.

DMARC (Domain-based Message Authentication, Reporting, and Conformance)

The policy layer. Tells receivers what to do when SPF or DKIM fails, and where to send reports.

_dmarc.example.com    TXT    "v=DMARC1; p=reject; rua=mailto:dmarc-aggregate@example.com; ruf=mailto:dmarc-forensic@example.com; pct=100; adkim=s; aspf=s"

Components:

  • p=: policy. none, quarantine, or reject.
  • rua=: aggregate reports (daily, summary). Always set this.
  • ruf=: forensic reports (per-message). Optional, can be high volume.
  • pct=: percentage of failing mail subject to the policy. Useful for gradual rollout.
  • adkim=, aspf=: alignment mode. s (strict), r (relaxed). Strict means From: domain must match exactly.

DMARC is the most important record. It's what makes spoofing your domain hard.

Pillar 2: Reputation

Mailbox providers (Gmail, Outlook, Yahoo) score every sender. Reputation drives delivery.

Reputation factors:

  • Authentication pass rates (SPF, DKIM, DMARC)
  • Engagement signals (opens, replies, marking as not-spam)
  • Negative signals (spam complaints, deletions without opens, blocking)
  • List hygiene (low bounce rates, no spam traps)
  • Volume consistency (sudden spikes look like spam)
  • Content patterns (link reputation, attachment patterns)
  • IP and domain history

Reputation is per (sending domain × mailbox provider). Gmail's view of you is independent of Outlook's.

Pillar 3: List quality and engagement

Authentication and reputation rest on list quality. Bad list = bad reputation eventually.

  • Only send to people who explicitly opted in
  • Confirmed (double) opt-in for marketing wherever feasible
  • Honor unsubscribes immediately and reliably
  • Remove hard bounces immediately
  • Sunset disengaged contacts (no opens in 6 months: reduce frequency or remove)
  • Avoid third-party lists, scraped emails, or "purchased opt-ins"

The single biggest deliverability lever for most senders is list hygiene.


Workflow

Step 1: Audit current state

Check the current DNS records:

dig +short txt example.com
dig +short txt selector1._domainkey.example.com
dig +short txt _dmarc.example.com

Also check:

  • Current bounce rate (target: under 2%)
  • Current spam complaint rate (target: under 0.1%)
  • Current open rate (varies by industry; falling trend is a warning)
  • Current sending volume

Tools: mxtoolbox.com, dmarcian.com, mail-tester.com (for individual messages).

Step 2: Fix authentication

If any of SPF, DKIM, DMARC is missing or misconfigured, fix first.

SPF fix order:

  1. Identify all legitimate senders (transactional ESP, marketing ESP, support tools, etc.)
  2. Get the include: value or IP for each
  3. Combine into a single SPF record
  4. Verify lookup count is under 10
  5. Use -all for hard fail (or ~all if rolling out gradually)

DKIM fix order:

  1. Generate a new selector per sending service
  2. Publish the public key in DNS
  3. Configure the ESP to sign with the private key
  4. Verify with a test send (check headers for dkim=pass)

DMARC fix order:

  1. Publish DMARC with p=none initially (monitoring mode)
  2. Set up an aggregate report endpoint (use a DMARC analytics service or your own)
  3. Watch reports for at least 2-4 weeks
  4. Identify any legitimate senders failing alignment; fix them
  5. Move to p=quarantine with pct=10, gradually increase
  6. Move to p=reject once confidence is high

The full progression typically takes 2-3 months. Rushing causes legitimate mail to bounce.

Step 3: Set up monitoring

Ongoing visibility:

  • DMARC aggregate reports: parsed daily. Watch for new sources, alignment failures, volume changes.
  • Bounce rate: per ESP dashboard. Target under 2% for transactional, under 5% for marketing.
  • Spam complaint rate: target under 0.1%.
  • Reputation tools: Google Postmaster Tools (free), Microsoft SNDS, Talos, Sender Score.
  • Blacklist monitoring: most ESPs include this; otherwise mxtoolbox.com or hetrixtools.com.

Step 4: Address list hygiene

  • Remove hard bounces from your sending lists immediately (most ESPs do this; verify)
  • Set up suppression lists (unsubscribes, manual blocks, deliverability sinkholes)
  • Sunset disengaged contacts (re-engagement campaigns or just remove)
  • Review opt-in flows: do people know what they're signing up for?

Step 5: Content and pattern audit

If reputation is good and authentication passes, check content:

  • HTML-only emails (no plain-text alternative) raise flags
  • Heavy image-to-text ratios raise flags
  • Suspicious link patterns (bare IPs, URL shorteners, hijacked domains)
  • Specific spam-trigger words in subject lines (less important than it used to be, still real)
  • Lots of mismatched domains in URLs
  • Generic from-name like "info@" rather than a real-sounding sender

Step 6: Plan for growth

Email volume affects reputation. Sudden spikes look like spam.

  • Warm up new sending IPs gradually (start at 50/day, 2x daily)
  • Warm up new domains gradually (similar pattern)
  • Spread sends throughout the day
  • Don't switch ESPs the week before a big launch

Step 7: Set up BIMI (optional but valuable)

BIMI (Brand Indicators for Message Identification) shows your logo next to authenticated emails in supporting clients (Gmail, Apple Mail, Yahoo, others).

Requires:

  • DMARC at p=quarantine or p=reject (so this comes after the DMARC progression)
  • A trademarked logo as an SVG (specific format requirements)
  • A Verified Mark Certificate (VMC) from a CA, for Gmail and others
  • A _bimi DNS record pointing to the SVG and VMC

BIMI improves trust signals and engagement. Worth doing once DMARC enforcement is in place.

Step 8: Document and revisit

Document the email architecture:

  • Sending domains and subdomains
  • Each ESP and what it sends
  • DNS records for each
  • Monitoring setup
  • Escalation if deliverability degrades

Revisit quarterly or when a new ESP is added.


Failure patterns

Multiple SPF records. Two or more SPF records on the same domain breaks SPF. Combine into one.

SPF DNS lookup limit exceeded. Too many include: directives or chained includes. Flatten or simplify.

DMARC at p=none forever. Monitoring without enforcement. Spoofing remains easy. Move to enforcement.

DMARC at p=reject too quickly. Legitimate mail bounces because alignment wasn't verified. Use the gradual rollout.

Sending from a different domain than the From: address. Causes alignment failures. Fix the From: domain or ensure proper alignment.

Using a shared ESP IP without ESP-specific configuration. Some ESPs don't sign with your DKIM by default; the signature is the ESP's, not yours. Configure custom DKIM.

Sending from a domain that also sends marketing. A spam complaint on a marketing email hurts transactional deliverability. Use a subdomain for transactional (transactional.example.com) or marketing (mail.example.com).

No bounce monitoring. Hard bounces accumulate, reputation tanks, deliverability cliff-falls. Monitor.

Bought lists or scraped contacts. Spam traps in those lists destroy reputation. Don't.

No double opt-in for marketing. Single opt-in lets bots and typos onto the list. Bots generate spam complaints, destroy reputation.

Ignoring DMARC reports. Reports show problems early. Set up a parser. Look weekly.

Treating "marked as not-spam" as the goal. The goal is to never land in spam in the first place. Once reputation is bad, recovery takes months.


Output format

A deliverability audit document includes:

  • Domain inventory: every sending domain
  • Authentication status: SPF, DKIM, DMARC per domain
  • Sender inventory: every ESP or service that sends mail, what's configured
  • Reputation status: per major mailbox provider where measurable
  • List hygiene: bounce rate, complaint rate, opt-in process
  • Findings: prioritized issues
  • Roadmap: SPF/DKIM/DMARC fixes, BIMI plan, monitoring plan
  • Monitoring setup: what's watched, where alerts go

Reference files

GitHub 저장소

rampstackco/claude-skills
경로: dist/codex/.agents/skills/email-deliverability
0
agent-skillsai-agentsanthropicclaudeclaude-aiclaude-code
FAQ

자주 묻는 질문

email-deliverability Skill이란 무엇인가요?

email-deliverability은(는) rampstackco이(가) 만든 Claude Skill입니다. Skill은 Claude가 필요할 때 불러오는 지침과 리소스를 묶어 추가 프롬프트 없이 email-deliverability 관련 작업을 수행할 수 있게 합니다.

email-deliverability은(는) 어떻게 설치하나요?

이 페이지의 설치 명령을 사용하세요. email-deliverability을(를) Claude Code 플러그인으로 추가하거나 저장소를 skills 디렉터리에 복제한 다음 Claude를 다시 시작해 Skill을 불러옵니다.

email-deliverability은(는) 어떤 카테고리에 속하나요?

email-deliverability은(는) 메타 카테고리에 속합니다.

email-deliverability은(는) 무료로 사용할 수 있나요?

네. email-deliverability은(는) AIMCP에 등록되어 있으며 무료로 설치할 수 있습니다.

연관 스킬

content-collections
메타

이 스킬은 콘텐츠 콜렉션(Content Collections)을 위한 프로덕션 검증된 설정을 제공합니다. 콘텐츠 콜렉션은 Markdown/MDX 파일을 Zod 검증이 포함된 타입 안전한 데이터 콜렉션으로 변환해주는 TypeScript 최우선 도구입니다. 블로그, 문서 사이트 또는 콘텐츠 중심의 Vite + React 애플리케이션을 구축할 때 타입 안전성과 자동 콘텐츠 검증을 보장하기 위해 사용하세요. Vite 플러그인 구성과 MDX 컴파일부터 배포 최적화 및 스키마 검증에 이르기까지 모든 것을 다룹니다.

스킬 보기
polymarket
메타

이 스킬은 개발자들이 Polymarket 예측 시장 플랫폼을 활용한 애플리케이션을 구축할 수 있도록 지원하며, 거래 및 시장 데이터를 위한 API 통합 기능을 포함합니다. 또한 WebSocket을 통한 실시간 데이터 스트리밍을 제공하여 실시간 거래와 시장 활동을 모니터링할 수 있습니다. 이를 통해 거래 전략을 구현하거나 실시간 시장 업데이트를 처리하는 도구를 생성하는 데 활용할 수 있습니다.

스킬 보기
creating-opencode-plugins
메타

이 스킬은 개발자들이 명령어, 파일, LSP 작업 등 25개 이상의 이벤트 유형에 연결되는 OpenCode 플러그인을 만들 수 있도록 돕습니다. JavaScript/TypeScript 모듈을 위한 플러그인 구조, 이벤트 API 명세, 구현 패턴을 제공합니다. OpenCode AI 어시스턴트의 라이프사이클을 사용자 정의 이벤트 기반 로직으로 가로채거나, 모니터링하거나, 확장해야 할 때 사용하세요.

스킬 보기
sglang
메타

SGLang은 RadixAttention 프리픽스 캐싱을 활용하여 JSON, 정규식, 에이전트 워크플로우를 위한 고속 구조화 생성에 특화된 고성능 LLM 서빙 프레임워크입니다. 특히 반복되는 프리픽스가 있는 작업에서 상당히 빠른 추론 속도를 제공하여 복잡한 구조화 출력 및 다중 턴 대화에 이상적입니다. 제약 디코딩이 필요하거나 광범위한 프리픽스 공유가 있는 애플리케이션을 구축할 때는 vLLM과 같은 대안보다 SGLang을 선택하십시오.

스킬 보기