comms-email-smtp-authentication-and-deliverability
Use for anything email: the federated versus proprietary divide that shapes the whole field, SMTP and email architecture, the retrieval protocols IMAP and POP, message format and MIME, SPF, DKIM and DMARC and how the three actually compose, deliverability and why legitimate mail lands in spam, and email encryption with S/MIME and PGP and why neither won. Includes the router for the whole communication technologies reference.
- License 许可
- MIT
- 最近更新
- 2026-11-15
- Stars
- ★ 5
- Forks
- 3
01 前言 Front Matter
- name
- comms-email-smtp-authentication-and-deliverability
- description
- Use for anything email: the federated versus proprietary divide that shapes the whole field, SMTP and email architecture, the retrieval protocols IMAP and POP, message format and MIME, SPF, DKIM and DMARC and how the three actually compose, deliverability and why legitimate mail lands in spam, and email encryption with S/MIME and PGP and why neither won. Includes the router for the whole communication technologies reference.
03 SKILL.md 正文
Communication Technologies: The Federated / Proprietary Divide, SMTP and Email Architecture, Retrieval Protocols, Message Format, SPF, DKIM and DMARC, Deliverability, and Email Encryption
Part 1 of 6 of the Communication Technologies reference (plugin
communication-technologies), covering §0–§7. Sibling skills:comms-telephony-pstn-ss7-voip-and-caller-id(§8–§12),comms-sms-rcs-signal-protocol-and-messaging-apps(§13–§18),comms-webrtc-video-conferencing-team-platforms-and-push(§19–§22),comms-encryption-metadata-interoperability-and-policy(§23–§27),comms-reference(§28–§33). Section numbers are shared across the set; a reference written as §N →skillpoints into that sibling skill.Currency: The protocols are old and stable. Two things moved. See §28 →
comms-referencefor cross-platform encrypted RCS, and the legal status of message scanning.
⚠️ The most-used software category on earth, built on protocols mostly designed before the threats they now face existed.
Builds on a cryptography reference (key exchange, forward secrecy, authentication), a wireless reference (cellular and Bluetooth layers), and a computer-hardware reference (networking). Complements a speaking-and-influence reference for the human side.
⚠️ GOTCHA boxes mark where the security or reliability property people assume is not the one the system actually provides.
The three ideas that organize this document:
- ⚠️ FEDERATED VERSUS PROPRIETARY IS THE DEEPEST DIVIDE (§1). Email, SMS and the phone network are federated — anyone can join, anyone can reach anyone, and nobody can fix them unilaterally. Signal, iMessage and Slack are proprietary — coherent, secure, improvable, and walled. Almost every frustration in this domain traces to that trade.
- ⚠️ "ENCRYPTED" IS FOUR DIFFERENT CLAIMS (§23 →
comms-encryption-metadata-interoperability-and-policy). Transport encryption, encryption at rest, end-to-end encryption, and end-to-end encryption with verified keys are wildly different guarantees, and marketing conflates them constantly.- ⚠️ METADATA IS USUALLY THE POINT (§24 →
comms-encryption-metadata-interoperability-and-policy). Who talked to whom, when, how often and from where survives content encryption — and for most surveillance purposes it is more valuable than content. A system can be perfectly end-to-end encrypted and still surveillable.
§0. Routing
| You want... | Go to |
|---|---|
| The federated/proprietary split | §1 |
| SMTP and email architecture | §2 |
| Retrieval protocols | §3 |
| Message format | §4 |
| ⚠️ SPF, DKIM, DMARC | §5 |
| ⚠️ Deliverability | §6 |
| Email encryption | §7 |
| PSTN | §8 → comms-telephony-pstn-ss7-voip-and-caller-id |
| ⚠️ SS7 | §9 → comms-telephony-pstn-ss7-voip-and-caller-id |
| VoIP and SIP | §10 → comms-telephony-pstn-ss7-voip-and-caller-id |
| VoLTE | §11 → comms-telephony-pstn-ss7-voip-and-caller-id |
| ⚠️ Caller ID and robocalls | §12 → comms-telephony-pstn-ss7-voip-and-caller-id |
| ⚠️ SMS | §13 → comms-sms-rcs-signal-protocol-and-messaging-apps |
| MMS | §14 → comms-sms-rcs-signal-protocol-and-messaging-apps |
| ⚠️ RCS | §15 → comms-sms-rcs-signal-protocol-and-messaging-apps |
| ⚠️ The Signal protocol | §16 → comms-sms-rcs-signal-protocol-and-messaging-apps |
| The messaging landscape | §17 → comms-sms-rcs-signal-protocol-and-messaging-apps |
| iMessage and FaceTime | §18 → comms-sms-rcs-signal-protocol-and-messaging-apps |
| ⚠️ WebRTC | §19 → comms-webrtc-video-conferencing-team-platforms-and-push |
| Video conferencing | §20 → comms-webrtc-video-conferencing-team-platforms-and-push |
| Team platforms | §21 → comms-webrtc-video-conferencing-team-platforms-and-push |
| Push and notifications | §22 → comms-webrtc-video-conferencing-team-platforms-and-push |
| ⚠️ What E2EE protects | §23 → comms-encryption-metadata-interoperability-and-policy |
| ⚠️ Metadata | §24 → comms-encryption-metadata-interoperability-and-policy |
| Interoperability regulation | §25 → comms-encryption-metadata-interoperability-and-policy |
| ⚠️ The encryption debate | §26 → comms-encryption-metadata-interoperability-and-policy |
| Emergency calling | §27 → comms-encryption-metadata-interoperability-and-policy |
| What's live | §28 → comms-reference |
| Misconceptions, numbers | §29–§30 → comms-reference |
| Sources, quick ref, method | §31–§33 → comms-reference |
§1. The Federated / Proprietary Divide
⚠️ FEDERATED ⚠️ email, SMS, the phone network, XMPP, Matrix
⚠️ ANYONE can participate · ⚠️ NOBODY can change it
unilaterally · ⚠️ security is limited by the WEAKEST
participant · ⚠️ evolution takes decades (§15, §28.1)
⚠️ AND THIS IS WHY EMAIL AND SMS ARE INSECURE — not
incompetence, but the cost of universal reach
⚠️ PROPRIETARY ⚠️ iMessage, Signal, WhatsApp, Slack, Discord
⚠️ Coherent security model · ⚠️ ships improvements in weeks ·
⚠️ walled garden · ⚠️ you depend on one company's continued
existence and goodwill
⚠️ ⚠️ THE PATTERN THAT KEEPS REPEATING: a proprietary system
demonstrates a capability, the federated standard adopts it
years later, badly. ⚠️ RCS versus iMessage is the current
instance and §28.1 is its resolution
⚠️ REGULATION IS NOW FORCING INTEROPERABILITY (§25) — ⚠️ an
attempt to get federation's reach with proprietary security,
and the jury is out
⚠️ ⚠️ THE HONEST TRADE ⚠️ you cannot have universal reach,
strong security, rapid evolution and no central authority
simultaneously. ⚠️ Every system here picks three
PART I — EMAIL
§2. SMTP and Email Architecture
⚠️ THE COMPONENTS ⚠️ MUA (your client) → MSA (submission) →
MTA (transfer, possibly several) → MDA (delivery) → mailbox
⚠️ SMTP ⚠️ from 1982, text-based, store-and-forward.
⚠️ Ports: 25 (MTA-to-MTA), 587 (⚠️ submission, authenticated),
465 (implicit TLS)
⚠️ ⚠️ THE ENVELOPE IS NOT THE HEADER. ⚠️ MAIL FROM and RCPT TO
are the SMTP envelope; the From: and To: you SEE are message
headers. ⚠️ THEY NEED NOT MATCH — and that gap is the
entire basis of email spoofing (§5)
⚠️ MX RECORDS in DNS route mail for a domain
⚠️ STARTTLS ⚠️ opportunistic encryption — ⚠️ and "opportunistic"
means STRIPPABLE by an active attacker unless MTA-STS or
DANE is in place, which most domains lack
⚠️ ⚠️ SMTP HAD NO AUTHENTICATION AT ALL BY DESIGN. ⚠️ Everything
in §5 is bolted on decades later, which is why it is a
layered mess of DNS records rather than a protocol feature
⚠️ BOUNCES ⚠️ hard vs soft · ⚠️ BACKSCATTER (bounces sent to a
forged sender) is a consequence of accepting-then-bouncing
§3. Retrieval Protocols
⚠️ POP3 downloads and traditionally deletes — ⚠️ a single-device model that predates people having several. ⚠️ IMAP keeps mail on the server with folders, flags and server-side search — ⚠️ the model that matches how people actually use email, and it is complex enough that implementations differ in maddening ways. ⚠️ JMAP is the modern replacement: ⚠️ JSON over HTTP, designed for mobile with efficient synchronization and push, and it is genuinely better — with the usual federated adoption problem (§1). ⚠️ Exchange/ActiveSync and Gmail's API are the proprietary equivalents, ⚠️ and the calendar and contacts integration is a large part of why organizations stay on them. ⚠️ The practical consequence for users: ⚠️ IMAP means your mail lives on a server someone else controls, which is the assumption §7 has to work around.
§4. Message Format
⚠️ RFC 5322 defines headers and body; ⚠️ MIME extends it to attachments, multiple character sets and multipart bodies — because the original format was US-ASCII text only. ⚠️ Multipart/alternative carries plain text and HTML versions of the same message, ⚠️ and HTML email is where tracking pixels, remote-image beacons and rendering inconsistency all live. ⚠️ Base64 and quoted-printable encode binary into the 7-bit-safe channel SMTP assumes — ⚠️ which is why attachments inflate roughly a third in transit. ⚠️ Message-ID, In-Reply-To and References are what make threading work, ⚠️ and clients that ignore them produce the broken threads everyone recognizes. ⚠️ Internationalized email (EAI/SMTPUTF8) allows non-ASCII addresses, ⚠️ and support remains patchy — a real equity issue for non-Latin-script users.
§5. ⚠️ SPF, DKIM, DMARC
⚠️ The three-layer patch over §2's missing authentication, and understanding what each actually checks is what makes email security legible.
⚠️ ⚠️ SPF ⚠️ a DNS record listing which IPs may send for a
domain. ⚠️ CHECKS THE ENVELOPE SENDER (MAIL FROM), NOT the
visible From: header
⚠️ THEREFORE SPF ALONE DOES NOT STOP SPOOFING of what the
user sees. ⚠️ It also BREAKS ON FORWARDING, because the
forwarder's IP is not in the original domain's record
⚠️ ⚠️ DKIM ⚠️ a cryptographic SIGNATURE over selected headers
and the body, with the public key in DNS.
⚠️ SURVIVES FORWARDING (as long as nothing modifies the
signed parts) · ⚠️ BREAKS when a mailing list appends a
footer or rewrites a subject
⚠️ ⚠️ DMARC ⚠️ THE ONE THAT MATTERS. ⚠️ Requires SPF or DKIM to
pass AND to be ALIGNED with the visible From: domain, and
publishes a POLICY: none, quarantine, or reject
⚠️ Plus aggregate and forensic REPORTING, which is how you
discover who is sending as you
⚠️ ⚠️ THE ALIGNMENT REQUIREMENT IS THE WHOLE POINT — it closes
the envelope-versus-header gap of §2
⚠️ ARC ⚠️ preserves authentication results across intermediaries
(mailing lists, forwarders) so DMARC does not destroy them
⚠️ BIMI displays a verified logo; ⚠️ requires DMARC enforcement
and usually a paid mark certificate — ⚠️ arguably a
compliance incentive dressed as a feature
⚠️ ⚠️ THE ROLLOUT TRAP ⚠️ going straight to p=reject without
reading reports first WILL silently drop legitimate mail from
systems you forgot about — the ticketing system, the payroll
provider, the marketing tool
§6. ⚠️ Deliverability
⚠️ Getting mail accepted is a different problem from sending it, ⚠️ and it is governed by reputation rather than by rules. ⚠️ What actually determines it: ⚠️ sending IP and domain reputation, authentication (§5), ⚠️ ENGAGEMENT SIGNALS (opens, replies, and especially deletions-without-reading), complaint rate, spam-trap hits, list hygiene, and consistency of volume.
⚠️ GOTCHA — complaint rate is the metric that kills you, and the thresholds are brutally low. ⚠️ Major providers publish complaint-rate targets in the region of a tenth of a percent — meaning a handful of "report spam" clicks per thousand messages is enough to damage a sending reputation. ⚠️ One easy unsubscribe link prevents more deliverability damage than any amount of content tuning.
⚠️ Bulk sender requirements now enforced by the large providers require ⚠️ SPF, DKIM, DMARC, one-click unsubscribe and complaint-rate limits — which formalized what was previously informal. ⚠️ Shared versus dedicated IPs, warming, and subdomain separation (⚠️ transactional mail on a different subdomain from marketing, so a bad campaign cannot take down your password resets). ⚠️ The uncomfortable structural point: ⚠️ a handful of providers effectively decide whose email is delivered, with no appeal — which is centralization arriving at a federated system through the back door (§1).
§7. Email Encryption
⚠️ Transport encryption (§2) protects the hop, not the message, ⚠️ and the provider reads everything. ⚠️ PGP/GPG — ⚠️ decentralized web of trust, no forward secrecy, and famously unusable for non-experts; ⚠️ the EFAIL research showed real vulnerabilities in how clients handled it. ⚠️ S/MIME — ⚠️ certificate-authority based, better organizational tooling, and therefore common in enterprise and government and rare elsewhere.
⚠️ GOTCHA — neither encrypts the SUBJECT LINE or the metadata (§24 →
comms-encryption-metadata-interoperability-and-policy). ⚠️ Who you emailed, when, and about what (per the subject) remains visible even with the body encrypted.
⚠️ Provider-based approaches (Proton, Tutanota) ⚠️ encrypt within their own systems and
fall back to links or plain mail outside — which is §1's trade in miniature.
⚠️ The honest assessment: ⚠️ email encryption has failed to achieve meaningful
adoption in thirty years, and if you need confidential messaging the practical answer is to
use something else (§16 → comms-sms-rcs-signal-protocol-and-messaging-apps, §17 → comms-sms-rcs-signal-protocol-and-messaging-apps).
PART II — TELEPHONY
04 同仓库更多 Skills
white-paper-writing
Use when writing white papers, technical reports, policy documents, thought leadership pieces, or any long-form authoritative document. Triggers on requests to write, structure, outline, or review white papers, technical reports, position papers, policy briefs, or evidence-driven arguments. Also use for B2B content marketing strategy, technical documentation requiring authority, academic-style reports, or any document that must educate while persuading. Use when someone asks "how should I structure this" for a long-form document.
prose-technical-mechanics
Use when editing or writing prose at a technical/mechanical level: choosing between Anglo-Saxon and Latinate diction, placing information in topic vs stress positions, writing periodic vs cumulative sentences, handling sentence rhythm, punctuation as logic, transitions, paragraph structure, or phonesthetics. Use when someone wants to understand WHY prose works or doesn't, not just surface-level style advice. Triggers on word choice, sentence structure, emphasis, rhythm, diction, or any deep writing mechanics question.
prose-mechanics
Use when writing, editing, or analyzing prose style; when trying to avoid generic AI-sounding writing; when applying specific prose styles such as journalistic, academic, narrative, KJV biblical, Hemingway plain, Latinate/Jamesian, Orwellian, Augustan, Victorian, Early Modern (Bacon, Browne, Milton), or free indirect discourse; or when coaching someone on writing voice and mechanics. Triggers on any request to improve, analyze, or write in a specific prose style, to fix "AI-sounding" writing, to make prose more vivid or concrete, to vary sentence rhythm, or to understand how a particular writer achieves their effects.
narrative-structure
Use when structuring stories, pitches, presentations, white papers, or any persuasive narrative; analyzing the hero's journey in fiction or non-fiction; applying StoryBrand for business communication; designing narrative arcs for thought leadership; or understanding Vogler, Snyder, Propp, or Campbell frameworks. Triggers on any narrative structure, storytelling, pitch structure, StoryBrand, hero's journey, three-act structure, Freytag, Save the Cat, narrative arc, or "how should I structure this story/pitch/presentation" request.
legal-brief-writing
Use when writing legal briefs, motions, or any court filing; when analyzing judicial cognition and persuasion; when applying elite brief-writing techniques from Roberts, Clement, Kagan, Garner; or when structuring legal arguments for maximum persuasion. Also use for autistic/neurodivergent advocates who need to understand how their natural communication style maps to elite legal advocacy. Triggers on any legal writing, brief drafting, motion preparation, or legal argument structuring request.
journal-publishability-criteria
Use whenever a research paper, article or preprint is being written, revised, reviewed or published, and specifically before every publication of docs/paper.md. Covers what legitimate journals actually gate on, in order of gating — the editor's desk screen (scope, novelty, package problems), what reviewers judge (methodological soundness, claims the evidence can carry, statistical and qualitative reporting), EQUATOR reporting guidelines, the ICMJE authorship criteria, COPE's view of text recycling, the near-universal generative-AI policy (no AI author, disclosure, author responsibility), TOP 2025 transparency standards, how to verify a journal is legitimate, submitting without an institution, and a pre-submission checklist. Triggers on requests to evaluate, review, prepare or submit a paper, to judge publishability, to write a declarations or data-availability section, to disclose AI use, or to check a journal.