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.
In this repository30 other files it gives its agents.
- AGENTS.mdAGENTS.md
- ad-to-landing-promise-matchSkill
- b2b-audience-signal-auditSkill
- cold-outbound-sequence-reviewSkill
- competitor-lead-angle-reviewSkill
- crm-lead-source-quality-auditSkill
- discovery-call-gap-analysisSkill
- disqualification-reason-minerSkill
- email-nurture-sequence-reviewSkill
- form-friction-finderSkill
- ghosted-after-the-demoSkill
- google-ads-lead-quality-auditSkill
- icp-reality-checkSkill
- landing-page-qualification-reviewSkill
- lead-magnet-offer-fitSkill
- lead-scoring-sanity-checkSkill
- lead-to-opportunity-drop-off-reviewSkill
- linkedin-ads-lead-quality-auditSkill
- meeting-prep-five-minutesSkill
- meta-ads-lead-quality-auditSkill
- objection-cheat-sheetSkill
- pipeline-hygiene-auditSkill
- proposal-win-loss-reviewSkill
- sales-follow-up-speed-auditSkill
- scale-readiness-check-lead-genSkill
- sdr-handoff-brief-generatorSkill
- weekly-b2b-lead-gen-readoutSkill
- what-to-say-to-this-companySkill
- won-deal-to-case-studySkill
- write-the-follow-upSkill
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
- Distinctive: line 14
Run this before anyone rewrites anything.
- Very specific: line 16
Do not use it to diagnose weak copy on a campaign that is demonstrably being delivered and opened.
- Distinctive: line 32
If someone hands over only a screenshot of a sequence dashboard, say what is missing and check what can be checked.
- Distinctive: line 37
Read what it returns and say plainly which of the three is missing, misconfigured or merely present but permissive.
- Distinctive: line 38
Check the domain age and reputation posture.
- Distinctive: line 40
Read the bounce rate as a list-quality signal first and a reputation signal second.
- Distinctive: line 41
Read the complaint rate as the strongest available proxy for whether recipients consider this spam.
- Distinctive: line 49
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 …
- Distinctive: line 50
If nothing is technically wrong, say that too, and hand the problem back to targeting and copy rather than inventing a deliverability cause.
- Distinctive: line 51
Never state a deliverability rate as measured unless a seed test or provider data supports it.
- Distinctive: line 91
Do not change DNS records.
- Distinctive: line 92
Do not promise an inbox placement rate.
- Distinctive: line 93
Do not treat open rate as proof of delivery.
- Distinctive: line 94
Do not recommend buying more domains as the default fix.
- Distinctive: line 95
Do not advise on cold outbound legality here.
≈ “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.
Files
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
- Run
../scripts/email_auth_check.pyover the records. Counting SPF lookups against the limit of ten, reading the qualifier onall, spotting DKIM test mode and tellingp=nonefromp=rejectare mechanical, and eyeballing them is how a domain passes an inspection and still lands in spam. Pass the records with--spf,--dkimand--dmarc, or a file with one per line. - Read what it returns and say plainly which of the three is missing, misconfigured or merely present but permissive. SPF that ends in
~allwith eleven includes, DKIM that does not verify, and DMARC set top=noneare three different problems with three different fixes. - 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.
- Compare volume per mailbox against a conservative ceiling. Most cold outbound problems are volume problems wearing a copy problem's clothes.
- 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.
- Read the complaint rate as the strongest available proxy for whether recipients consider this spam. It matters more than any other number here.
- 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.
- 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
| Record | Present | Correct | Risk | Fix |
| 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 problemMissing 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.
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
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.

