Follow the assessment, not just the pass label
Forwarded Email Authentication Without a Blanket Verdict
A message reaches a shared alias, passes through another service, and arrives in a personal inbox with a different authentication story. One service reports SPF pass and another reports fail; DKIM may still be reported as passing. Those observations need not contradict each other. They can describe different deliveries or different identities. Start by identifying the received copy and the service making each declaration, then investigate what the intermediary actually changed.
Use the Email Header Analyzer to inspect authorized raw headers from the delivered copy. Paste them into the raw-header textarea, then choose Analyze headers explicitly. Load example only fills public sample text, and Clear removes the current input and results. This is a local parser with no API calls or DNS lookups, not a live receiver test or an authentication-verification service.
Establish which delivered message you are inspecting
Distinguish automatic forwarding from a person pressing Forward. An automatic relay can preserve much of the original message while creating another transport transaction. A manual forward commonly creates a new outer message containing quoted material from an earlier one. The outer message has its own headers, and a typed From or Sent line in the body is not a raw delivery record for the earlier message. Ask which mailbox received the copy under investigation and obtain that copy’s actual header block. Do not start by trying to explain an original sender’s result using only a colleague’s newly composed forward.
In browser Gmail, start with the message menu and Show original. In current Outlook interfaces, use the appropriate message-details or Internet-headers view. Follow the Gmail headers guide or Outlook headers guide for the relevant client. A screenshot of an authentication badge is not a substitute for the underlying fields when the question concerns multiple services. Keep a private note identifying the receiving mailbox, collection time, and whether the copy came through a known alias or forwarding rule. That context is yours to establish; the analyzer does not connect to accounts or discover their rules. If two staff members received separate copies, label them separately even when their subjects and visible contents match. Similar appearance does not establish an identical delivery path or assessment.
Paste header text rather than a complete conversation. The first blank line ends the header block, and body content after it is ignored, even if it resembles headers. The analyzer neither renders the body nor reads attachments or .eml files. Before analysis, respect the limits of 128 KiB UTF-8 source, 2,000 lines, 500 fields, 128 Received fields, and 32 fields across Authentication-Results and ARC-Authentication-Results combined. Each value separately must be no longer than 8,192 UTF-16 units, with its retained whitespace and folded line breaks included. Oversized or structurally malformed input is rejected atomically, including orphan continuations and non-header lines before the separator. A rejection is a source problem, not evidence that forwarding caused an authentication failure or that a sender is malicious.
Explain SPF in terms of the new transport hop
SPF evaluates whether a connecting system is authorized for the transport identity being checked. It does not directly authenticate the friendly display name or every address in the message. Consider a synthetic route from sender.example through forwarder.example to receiver.example. At the final receiver, the connecting system may now be the forwarder rather than the original sending service. If the original transport identity remains in use, the authorization question is different from the first receiver’s question. A reported failure can therefore be consistent with forwarding, but consistency is a hypothesis, not proof that a particular failure is harmless.
Some forwarding arrangements rewrite the envelope sender, including schemes designed to route delivery errors back through the forwarder. That can produce a different Return-Path and a different SPF identity at the next receiver. Do not equate a pass for the forwarding domain with a pass for the original visible From domain. Write down the exact declared smtp.mailfrom or smtp.helo property when present, and leave an absent property unknown. The From, Reply-To, and Return-Path guide separates those address roles. Different addresses alone do not prove spoofing, and rewriting a transport address does not establish ownership of the author mailbox.
DMARC adds an alignment question involving the visible From domain and an authenticated SPF or DKIM identity. The presence of any SPF pass somewhere in the source is not enough to answer that question. Read a declared DMARC result alongside the identities in the same assessment rather than assembling a new verdict from unrelated fields. The authentication results guide explains the methods separately. This analyzer does not query current policies, reconstruct the original SMTP exchange, or calculate a verified alignment outcome. Do not change DNS records or add a forwarding service to an allowlist solely because a pasted result seems inconvenient.
Ask what happened to the material protected by DKIM
DKIM and SPF respond to different aspects of forwarding. A DKIM signature can survive another transport hop if the material it protects remains compatible with verification. A changed connecting IP address is not, by itself, a change to signed content. Conversely, a forwarding or mailing-list service may alter a protected Subject, add a footer to the body, or re-encode content in a way that affects a signature. Google’s current forwarding guidance highlights the distinction between preserving DKIM-protected material and changing the delivery path. Do not assume that every forward breaks DKIM or that every unchanged-looking message retains a valid signature.
Inspect the declared d= signing domain, s= selector, and a= algorithm as identifiers, not a locally verified success. A message can contain several DKIM-Signature fields, including a signature added by an intermediary. A later signature from forwarder.example is not automatically evidence that an earlier signature from sender.example survived. Keep each reported result attached to its declared identity and assessment service. The analyzer extracts supported declarations but performs no cryptographic verification, key retrieval, body-hash validation, or DNS lookup. A plausible selector name cannot substitute for those operations, and the displayed algorithm does not prove that the signature was checked.
For a support investigation, ask the forwarding administrator which transformations are configured and whether an approved before-and-after comparison is available. Do not paste the email body into the analyzer to test a footer theory: body content is outside its analysis. Do not edit the headers to remove a list tag and claim that you have repaired DKIM. Changing a local copy does not resend or reauthenticate the original delivery. If the only available evidence is the final received copy, say so. A narrow finding such as possible content modification is more useful than naming a specific broken transformation without access to the relevant configuration or logs. Ask whether the administrator means an unchanged original signature or a newly added intermediary signature when describing a successful check. Those findings answer different questions about the route.
Separate each assessment
Inspect the delivered copy
Read each service declaration in context before connecting a result change to a forwarding hop.
Keep authentication reports and ARC assertions distinct
Read all Authentication-Results fields, preserving their source order and their individual authserv-id values. An earlier service might report a pass before forwarding, while a later service reports a failure after a different transport connection or content change. Those reports should not be merged into one green or red status. Neither the first pass nor the last fail automatically wins. A familiar service identifier is a claim in the text, and position alone cannot prove a trusted boundary. The analyzer does not decide which report belongs to your genuine receiving infrastructure. That requires provider knowledge and, when appropriate, administrator evidence beyond the pasted source.
ARC can carry intermediary authentication assessments using ARC-Authentication-Results, ARC-Message-Signature, and ARC-Seal fields. Instance tags associate declarations with an ARC set; a seal can include a declared chain-validation value such as cv=pass. Seeing those strings is not the same as validating the chain or trusting the sealer. The analyzer presents ARC separately from direct authentication and does not perform ARC cryptography. Do not promote a declaration inside an earlier ARC assessment into a current SPF or DKIM pass. ARC can provide context for an indirect delivery, but it is not a universal authentication pass, identity check, or safety label for the message.
Microsoft’s current ARC guidance distinguishes a service’s presence from recipient-side decisions about trusted sealers. That distinction matters even when a chain is valid in a capable receiving system: acceptance still depends on the receiver’s policy and confidence in the intermediary. Do not configure trust from an unfamiliar domain name pasted into a support conversation. Ask the authorized administrator whether the intermediary is expected, required, and already part of the documented mail path. An ARC declaration can help formulate that question without answering it. No result in this local parser guarantees inbox placement, proves a message is not spam, or independently authenticates the person requesting a response.
Read the route without sorting away inconsistencies
Use Received declarations to look for context around a suspected forwarding hop, not as an independently observed map. The analyzer reverses their source order for an oldest-to-newest reading view; it does not sort them chronologically. That preserves inconsistencies that timestamp sorting could hide. A familiar host label is still a declaration, not a successful DNS check or proof of a connection. Relate a suspected intermediary to known forwarding settings outside the tool. The Received headers and delivery delays guide explains how to describe the route without treating every line as trusted evidence of where the message physically traveled.
Adjacent timestamp deltas are signed differences between parseable declarations in that reversed-source sequence. If one declaration reads 10:05 and the following one reads 10:03 after timezone interpretation, the negative difference is a clock or order inconsistency to investigate. It is not evidence of a message traveling backward in time or a negative queue duration. An unknown timestamp keeps the adjacent difference unknown; do not replace it with the Date field or a neighboring interval. Nor should you add every positive interval and call the sum the verified forwarding delay. Delivery timing needs a reliable observation point and appropriate server records.
For a direct-versus-forwarded comparison, collect each delivered copy separately and label its context privately before beginning. Analyze one, note a narrow nonsecret observation in your own approved workspace, clear it, then analyze the other. The analyzer provides no saved comparison, projects, history, storage, export, PDF, or clipboard control. A difference in results can reflect different deliveries, not a change to the same stored message. Keep source omissions explicit: if the direct copy is unavailable, the comparison cannot establish its earlier result. Do not manufacture a baseline by removing the forwarder’s fields from the final received source and treating the remainder as an untouched earlier delivery. Record whether the expected forwarding rule was confirmed by its owner or merely inferred from a hostname. That distinction keeps configuration knowledge separate from the declarations being inspected.
Hand off a focused question without weakening mail controls
A useful support note identifies the receiving copy, known forwarding arrangement, service-specific authentication declarations, and the remaining question. For example, ask whether an expected intermediary rewrites the envelope identity or alters protected content. Do not attach a blanket recommendation to disable DMARC, trust every ARC sealer, or bypass filtering. Preserve the original privately. If another person needs an illustration, use the safe header-sharing guide to prepare a manually redacted copy outside the analyzer and list changed fields. Redaction can remove evidence needed for a definitive investigation, so agree on the required detail with the authorized recipient first.
The genuine isolated panel uses standalone PHP, an opaque iframe origin, no-store, and connect-src 'none'. Analysis runs in a terminable local worker with a six-second deadline. Clear, hidden-page cleanup, pagehide, pageshow, and freeze wipe the input and derived result DOM and stop active work. Returning to the page does not restore the previous case. A timeout means analysis did not complete, not that a complicated route passed authentication. These controls do not guarantee resistance to extensions, a compromised device, or a hostile parent replacing the panel, and do not securely erase all browser memory or copies made elsewhere.
Keep the operational conclusion proportional to the evidence. Forwarding is a possible explanation for changed authentication results, not an automatic excuse for every failed result or suspicious instruction. The analyzer cannot prove a trust boundary, verify a signature or identity, classify spam, guarantee deliverability, connect an account, or fix a mail rule. Finish by clearing the session and taking the precise question to the person responsible for the relevant mail system. The Email Header Analyzer overview brings the header-reading tasks together; Tools lists separate utilities without transferring your message into another service or promising a mail-security verdict.
Questions about Forwarded Email Authentication
Can SPF fail after forwarding while DKIM is reported as passing?
Yes. Forwarding can change the connecting system used for an SPF assessment while leaving DKIM-protected material intact. Read the declared identity and assessment service for each result. This analyzer reports the supplied declarations; it does not verify that either result is genuine or correct.
Does ARC pass mean the forwarded message is safe?
No. A declared ARC pass is not local chain verification, trust in every intermediary, or direct authentication of the author. ARC can carry earlier assessments for a receiver to consider under its own policy. It does not guarantee safety, spam status, or delivery.
Can I analyze the original email from text quoted below a forwarded message?
No. Quoted body text is not the original raw delivery-header block. The parser stops at the first blank line and ignores the body. Obtain authorized raw headers from the relevant delivered copy; the tool does not open email files, read attachments, or connect to a mailbox.
Continue with another email-header guide
Interpret a supplied header in the Email Header Analyzer or choose a related task below. Explore more practical utilities in all free tools.
Evidence before a mail-rule change
Review forwarding declarations carefully
Analyze one authorized header block locally, preserve unresolved questions, and clear the session afterward.