Failed controls that can make the message ineligible or unsafe to send.
All tools
AUTHENTICATION / PERMISSION / MESSAGE / OPERATIONSScore the evidence.
Score the evidence.
Expose the risk.
Inspect a final message and its operating signals without turning a heuristic into an inbox promise.
- Transparent point ledger
- Raw-header inspection
- Ordered repair queue
SCORE97/100
STRONG MECHANICAL EVIDENCE
The fixture clears this model's mechanical controls.
This score describes the pasted fixture and declared operating conditions. It does not measure private reputation, guarantee acceptance, or predict inbox placement.
Evidence coverage85%
14 observed · 8 declared · 0 missing
Warnings and unknowns that require evidence or a deliberate decision.
links.northstar.example
Encoded MIME size and provider transformations can be larger.
Score composition
Applicable points onlyEvidence and deduction ledger
Identity · Passed
SPF result from the received route
Authentication-Results reports spf=pass.
6/6
Identity · Passed
DKIM signature validates
Authentication-Results reports dkim=pass.
7/7
Identity · Passed
DMARC passes with visible-From alignment
Authentication-Results reports dmarc=pass.
9/9
Identity · Passed
Visible From identity is parseable
Observed From domain: news.northstar.example.
4/4
Identity · Passed
Forward and reverse DNS evidence
Declared verified by the operator.
2/2
Operations · Passed
TLS observed on the delivery route
Declared verified by the operator.
2/2
Permission · Passed
Recipient relationship supports this traffic
Declared confirmed opt-in with explicit recipient action.
8/8
Permission · Passed
Visible body unsubscribe
A visible unsubscribe or preference link was found in the rendered HTML.
5/5
Permission · Passed
One-click unsubscribe headers
List-Unsubscribe contains HTTPS and List-Unsubscribe-Post declares One-Click.
6/6
Permission · Passed
Unsubscribe headers covered by DKIM
The parsed DKIM h= list includes both unsubscribe fields.
3/3
Permission · Passed
Suppression propagation is verified
Declared: opt-outs and terminal failures stop future sends.
3/3
Message · Passed
Core message fields are present
Date, Message-ID, MIME-Version, and From were found.
5/5
Message · Passed
Plain-text and HTML alternatives are signaled
Content-Type declares multipart/alternative.
3/3
Message · Passed
Links use reviewable destinations
3 links across 1 absolute domains; no narrow mechanical link warning.
5/5
Message · Passed
Images have alternative text
All 1 images include an alt attribute.
3/3
Message · Passed
Copy avoids mechanical deception signals
No narrow mechanical copy warning was triggered.
3/3
Message · Passed
Display name identifies the sender
Observed display name: Northstar Updates.
3/3
Message · Passed
HTML payload stays reviewable
688 B of HTML was pasted.
3/3
Operations · Review
Spam complaint rate is controlled
Declared rate: 0.12%; below 0.3% but above the model's full-credit range.
Define the receiver-specific numerator and denominator, stop risky sources, and suppress complainants.4/7
Operations · Passed
Hard-bounce rate is controlled
Declared rate: 0.42%.
4/4
Operations · Passed
Volume change has a controlled ramp
Declared: a planned ramp is in use.
4/4
Operations · Passed
Complaint signals have an owner
Declared: complaint telemetry is monitored.
3/3
Receiver requirement lens
Scope-aware, not placement predictionsGmailPersonal Gmail · high-volume lens
Mechanical requirements appear represented.
The lens checks SPF, DKIM, DMARC alignment, low complaints, and subscription one-click mechanics for the declared high-volume profile.
Google's scope and private enforcement signals remain receiver-controlled.Yahoo / AOLBulk sender lens
The fixture covers the scored sender controls.
The lens separates all-sender authentication and DNS basics from bulk authentication and marketing unsubscribe controls.
Yahoo does not publish a universal volume threshold for its bulk classification.Outlook.comConsumer domains · high-volume lens
The scored authentication trio passes.
For the declared high-volume profile, this lens checks SPF, DKIM, and DMARC results from the pasted received message.
Acceptance and placement still depend on Microsoft's broader filtering and sender history.Ordered repair queue
1 actions- 01Spam complaint rate is controlledReview
Define the receiver-specific numerator and denominator, stop risky sources, and suppress complainants.
HANDOFF, NOT CERTIFICATION
Keep the score, evidence boundary, and next actions together.
The manifest records every point and declared input so another operator can reproduce the result. Receiver reputation and placement remain outside this browser-only model.
Read the full agent guideHOW TO USE THE NUMBER
A score is a triage instrument, not a receiver verdict.
The model rewards observable message evidence and clearly marked operating declarations. It cannot see private reputation systems, recipient-level engagement, historical traffic, current blocklists, provider configuration, DNS state, or the final SMTP conversation.
Use it to find missing controls and preserve a repeatable handoff. Use delivered fixtures, provider telemetry, receiver portals, DNS traces, and SMTP responses to establish what happened in production.
- 01Receive
Send through the real production route and copy headers from the received message.
- 02Declare
Record traffic purpose, permission, DNS/TLS evidence, complaints, bounces, and volume change.
- 03Inspect
Review authentication, alignment, unsubscribe, structure, links, and message-size deductions.
- 04Repair
Resolve failed identity and recipient-rights controls before cosmetic content warnings.
- 05Verify
Resend the final build and archive receiver-specific evidence with the revision.
SCORING QUESTIONS
What does the score leave out?
Anything the browser cannot establish from the pasted fixture or your explicit declarations remains outside the claim.
Does 100 mean the email will reach the inbox?+
No. It means the applicable checks in this limited model earned all available points. Receivers still evaluate reputation, recipient behavior, traffic history, content, security signals, and context this tool cannot observe.
Why is evidence coverage separate from the score?+
A plausible declaration and a received-message observation are not equally strong. Coverage shows how much of the result comes from observable headers or content versus declarations or missing evidence.
Why can authentication failures outweigh content warnings?+
Authentication and alignment can determine eligibility under receiver requirements and protect identity. Rewriting a phrase cannot compensate for failed sender identity or broken opt-out mechanics.
Should transactional mail include one-click unsubscribe?+
Not universally. The tool marks subscription-only unsubscribe controls not applicable for a transactional profile. Classification must reflect the real purpose and recipient expectation, not a label chosen to avoid a control.
Is 0.3% a safe complaint target?+
No. Gmail and Yahoo publish 0.3% as an important upper boundary in specified contexts, not a healthy target. The scorer reserves full points for a lower declared rate and still requires receiver-specific denominator definitions.
Can pasted headers contain sensitive data?+
Yes. Raw headers can include recipient addresses, message identifiers, routing details, and unsubscribe tokens. Redact unnecessary personal data and secrets before exporting or sharing the result.