agentsclimarketplace

Email deliverability debugger

Skill ComeOnOliver/skillshub/skills/TerminalSkills/skills/email-deliverability-debugger

🧠 The right skill, one API call. AI agent skills registry with token-efficient skill resolution. 5,000+ skills from 500+ top repos.

Install
npx -y skills add ComeOnOliver/skillshub --skill email-deliverability-debugger

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

SKILL.md

5.3 KB, ~1.4k tokens by cl100k_base, as published. Nobody here has run it

Email Deliverability Debugger

Overview

This skill systematically diagnoses why emails fail to reach inboxes by checking the entire delivery chain: DNS authentication (SPF, DKIM, DMARC), sending IP reputation, email content, headers, and provider-specific configuration. It produces actionable fixes ranked by impact.

Instructions

Step 1: Check DNS Authentication Records

# SPF record
dig +short TXT example.com | grep "v=spf1"

# DKIM (check common selectors)
for sel in s1 s2 k1 google default sendgrid; do
  dig +short TXT ${sel}._domainkey.example.com
done

# DMARC
dig +short TXT _dmarc.example.com

Validate each:

  • SPF: Must include all sending services, ≀10 DNS lookups, end with -all (hardfail)
  • DKIM: Each sending service needs its own selector with valid key
  • DMARC: Should have p=quarantine or p=reject; p=none provides no protection

Step 2: Count SPF Lookups

SPF has a hard limit of 10 DNS lookups. Count each include:, a:, mx:, and redirect= as 1 lookup. Nested includes count too.

# Check SPF record and count includes
dig +short TXT example.com | grep spf
# Then recursively check each include
dig +short TXT _spf.google.com
dig +short TXT sendgrid.net

If at or near 10, recommend SPF flattening or removing unused services.

Step 3: Analyze Email Headers (if provided)

Key headers to check:

  • Authentication-Results: Shows SPF, DKIM, DMARC pass/fail
  • X-MS-Exchange-Organization-SCL: Microsoft spam confidence (0-9, >5 = spam)
  • X-Spam-Score: Generic spam score
  • Received-SPF: SPF result with detail
  • DKIM-Signature: Which domain signed and which selector used
  • Return-Path: Must align with From domain for SPF alignment

Step 4: Check IP Reputation

Query major blacklists:

# Reverse IP for DNSBL queries (e.g., 1.2.3.4 becomes 4.3.2.1)
REVERSED="4.3.2.1"
for bl in zen.spamhaus.org b.barracudacentral.org bl.spamcop.net; do
  dig +short ${REVERSED}.${bl}
done
# Non-empty result = listed on that blacklist

Common blacklists to check: Spamhaus, Barracuda, SpamCop, SORBS, CBL, URIBL.

Step 5: Check Content Issues

Flag common spam triggers:

  • Subject line with ALL CAPS or excessive punctuation (!!!)
  • HTML-only email without text/plain alternative
  • High image-to-text ratio
  • URL shorteners (bit.ly, etc.) in email body
  • Missing unsubscribe header (required for bulk, good practice for all)
  • Missing List-Unsubscribe header

Step 6: Produce Fix Report

Organize by priority:

  1. Critical: Authentication failures (SPF/DKIM/DMARC fail)
  2. High: IP blacklisting
  3. Medium: Content issues, missing headers
  4. Low: Best practice improvements

Include exact DNS records to add/modify and provider-specific configuration steps.

Examples

Example 1: Full Deliverability Audit

User prompt:

Our emails to Gmail users are going to spam. Domain: notifications.myapp.com

Agent output:

Deliverability Audit β€” notifications.myapp.com

❌ CRITICAL: DMARC alignment failure
   From: notifications.myapp.com
   Return-Path: bounces.sendgrid.net (different domain)
   β†’ SPF passes for sendgrid.net but doesn't align with From domain
   Fix: Configure SendGrid authenticated domain so Return-Path uses your domain

⚠  HIGH: DMARC policy too weak
   Current: v=DMARC1; p=none
   β†’ Gmail honors DMARC but p=none means failures aren't acted on
   Fix: v=DMARC1; p=quarantine; rua=mailto:[email protected]; pct=100

βœ… SPF: Valid (7 lookups)
βœ… DKIM: Valid (s1 selector, 2048-bit)
βœ… IP reputation: Clean on all major blacklists

Priority fix order:
1. Configure SendGrid domain authentication (fixes Return-Path alignment)
2. Update DMARC to p=quarantine
3. Monitor DMARC reports for 2 weeks before moving to p=reject

Example 2: Specific Bounce Diagnosis

User prompt:

Getting 550 5.7.1 rejections from outlook.com. What's wrong?

Agent output:

550 5.7.1 from Outlook = message rejected due to content or policy

Diagnosis checklist:
1. βœ… SPF passes for your domain
2. ❌ DKIM signature uses d=sendgrid.net, not your domain
   β†’ Microsoft strict mode rejects misaligned DKIM
3. ❌ Your sending IP 149.72.x.x is on Spamhaus PBL
   β†’ This is a policy listing, not spam β€” means IP is flagged as dynamic/cloud

Fixes:
1. Set up DKIM custom domain in SendGrid (Settings β†’ Sender Authentication)
   Add CNAMEs: s1._domainkey.yourdomain.com β†’ s1.domainkey.u1234.wl.sendgrid.net
2. Request Spamhaus PBL removal at spamhaus.org/pbl/query/
3. After fixes, use mail-tester.com to verify score >8/10 before bulk sending

Guidelines

  • Check authentication FIRST β€” 80% of deliverability issues are DNS authentication failures
  • Alignment matters more than pass/fail β€” SPF can pass but still fail DMARC if domains don't align
  • Separate transactional from marketing β€” different subdomains and IPs protect transactional reputation
  • DMARC reports are gold β€” always recommend setting up rua aggregate reports
  • Don't recommend p=reject immediately β€” go none β†’ quarantine β†’ reject over 4-6 weeks while monitoring
  • IP reputation recovers slowly β€” blacklist removal is quick, but reputation rebuilding takes 2-4 weeks of clean sending
  • Test with real recipients β€” mail-tester.com, Gmail postmaster tools, and Microsoft SNDS are free and essential

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.