agentleFS
mardab96mardab96/b2b-lead-generation-claude-skills/spam-folder-check/SKILL.md

spam-folder-check

Skill5 starsChanged 4 months agoMITView on GitHub

Checks whether cold emails are actually reaching inboxes by reading SPF, DKIM and DMARC records, sending volume, bounce and complaint rates. Use when reply rates fall without the copy changing, when a new domain starts sending, or when someone asks whether their emails are going to spam.

The file

Show its rules and text

We split this skill into the rules it gives an agent

15 rules. About 0 are common, 14 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈14

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Run this before anyone rewrites anything.

    line 14
  2. Very specific:

    Do not use it to diagnose weak copy on a campaign that is demonstrably being delivered and opened.

    line 16
  3. Distinctive:

    If someone hands over only a screenshot of a sequence dashboard, say what is missing and check what can be checked.

    line 32
  4. Distinctive:

    Read what it returns and say plainly which of the three is missing, misconfigured or merely present but permissive.

    line 37
  5. Distinctive:

    Check the domain age and reputation posture.

    line 38
  6. Distinctive:

    Read the bounce rate as a list-quality signal first and a reputation signal second.

    line 40
  7. Distinctive:

    Read the complaint rate as the strongest available proxy for whether recipients consider this spam.

    line 41
  8. Distinctive:

    If the main company domain is being used for cold outbound, say so plainly: the downside is the entire company's email, not just the …

    line 49
  9. Distinctive:

    If nothing is technically wrong, say that too, and hand the problem back to targeting and copy rather than inventing a deliverability cause.

    line 50
  10. Distinctive:

    Never state a deliverability rate as measured unless a seed test or provider data supports it.

    line 51
  11. Distinctive:

    Do not change DNS records.

    line 91
  12. Distinctive:

    Do not promise an inbox placement rate.

    line 92
  13. Distinctive:

    Do not treat open rate as proof of delivery.

    line 93
  14. Distinctive:

    Do not recommend buying more domains as the default fix.

    line 94
  15. Distinctive:

    Do not advise on cold outbound legality here.

    line 95

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Spam Folder Check

Use this skill when

Someone suspects their emails are not arriving, and nobody has checked.

The tell is almost always the same: reply rates drop, the copy did not change, the list did not change, and the team's first instinct is to rewrite the subject line. Subject lines do not matter to a message that never renders in an inbox.

Run this before anyone rewrites anything.

Do not use it to diagnose weak copy on a campaign that is demonstrably being delivered and opened. That is cold-outbound-sequence-review.

Required input

At minimum, the sending domain. That alone gets you the authentication picture.

Better, add whatever of this exists:

  • the DNS TXT records for SPF, DKIM and DMARC, pasted, or permission to read them
  • sending volume per day per mailbox, and how long the domain has been sending
  • bounce rate, split into hard and soft if the tool reports it
  • spam complaint rate
  • whether the sending domain is the main company domain or a lookalike bought for outbound
  • the sending tool and whether mailbox warmup is switched on
  • open rate history, with the caveat below

If someone hands over only a screenshot of a sequence dashboard, say what is missing and check what can be checked.

Analysis workflow

  1. Run ../scripts/email_auth_check.py over the records. Counting SPF lookups against the limit of ten, reading the qualifier on all, spotting DKIM test mode and telling p=none from p=reject are mechanical, and eyeballing them is how a domain passes an inspection and still lands in spam. Pass the records with --spf, --dkim and --dmarc, or a file with one per line.
  2. Read what it returns and say plainly which of the three is missing, misconfigured or merely present but permissive. SPF that ends in ~all with eleven includes, DKIM that does not verify, and DMARC set to p=none are three different problems with three different fixes.
  3. Check the domain age and reputation posture. A domain that started sending in volume three weeks after registration behaves differently from one sending for four years, and the fix is patience, not configuration.
  4. Compare volume per mailbox against a conservative ceiling. Most cold outbound problems are volume problems wearing a copy problem's clothes.
  5. Read the bounce rate as a list-quality signal first and a reputation signal second. A hard bounce rate above a few percent is a list that was never verified, and it damages the domain for months after the campaign ends.
  6. Read the complaint rate as the strongest available proxy for whether recipients consider this spam. It matters more than any other number here.
  7. Separate what is broken from what is merely unproven. Open rates cannot confirm delivery. Apple Mail Privacy Protection, on by default since 2021, loads the tracking pixel on delivery whether or not anyone reads the message, so on an Apple-heavy list the number is inflated by an unknown amount. Do not build a delivery conclusion on it.
  8. Name the one fix that comes first, and say what it will not fix.

Decision rules

  • Authentication failures outrank everything. There is no point tuning volume on a domain that cannot prove who it is.
  • A brand new domain sending at volume is a warmup problem, and no configuration change substitutes for time.
  • If the main company domain is being used for cold outbound, say so plainly: the downside is the entire company's email, not just the campaign.
  • If nothing is technically wrong, say that too, and hand the problem back to targeting and copy rather than inventing a deliverability cause.
  • Never state a deliverability rate as measured unless a seed test or provider data supports it. Everything else is inference.

Output format

Verdict

One line: are the emails likely arriving, likely filtered, or unknowable from what was supplied.

Authentication

RecordPresentCorrectRiskFix
SPF
DKIM
DMARC

Sending posture

Domain age, volume per mailbox, warmup status, bounce rate, complaint rate, each with a plain read.

Fix order

Numbered, first fix first, with what each one will and will not solve.

End with:

  • Decision: fix now / warm up and wait / stop sending / this is not a deliverability problem
  • Missing data: only what would change the answer
  • Where a seed test would settle the question, say so

Practical example

Someone says replies fell from a steady trickle to nothing over two weeks, and nothing about the campaign changed.

The check finds DMARC absent, SPF technically valid, DKIM signing correctly, the domain four months old, and volume raised from 30 to 120 emails per mailbox per day eleven days ago.

The answer is not the missing DMARC record, even though it should be added. The answer is that the volume quadrupled on a young domain, and the fix is to drop back below the previous level, hold it, and stop treating send volume as a growth lever on a domain this young. The DMARC fix goes second, because it protects the domain rather than rescuing this campaign.

Guardrails

  • Do not change DNS records. Produce the record to add and hand it to whoever owns the domain.
  • Do not promise an inbox placement rate. Nobody outside the receiving provider can measure it directly.
  • Do not treat open rate as proof of delivery.
  • Do not recommend buying more domains as the default fix. It hides the cause and multiplies it.
  • Do not advise on cold outbound legality here. Deliverability and lawfulness are separate questions and this skill answers only the first.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.