One install.
Fully autonomous email.

What we enforce on every send. What we reject. How you can verify.

Version v1.0 · Published 2026-04-18

This section describes what Outboxr enforces on every send delivered through every domain in its network. We treat it as a binding commitment, not marketing. Every claim ends with a way for you to verify it yourself.

What we are

Outboxr composes newsletters and 1:1 outreach, routes them through warmed sender identities in a trusted network, and auto-heals reputation based on postmaster and ESP feedback. We are AI-assisted. We disclose it on every send that goes out at scale.

What we reject, by name

• ·Hallucinated URLs in sends. Stripped pre-delivery via sender-owned allowlist.
• ·Unverifiable factual claims. Extracted and checked by a second-pass model. Low-confidence claims block delivery.
• ·Purchased or scraped lists. Recipients without verifiable consent are refused at admission. Permission is checked per batch.
• ·Spam-farm footprints. Shared sender identity, duplicated voice, synchronized send windows, or shared sending IPs across unrelated domains. Blocked at admission. Re-audited weekly.
• ·Deceptive subject lines or headers. CAN-SPAM violations, fake thread headers, spoofed From addresses, misleading unsubscribe links. Non-negotiable.
• ·Missing or broken unsubscribe. Every send carries a working one-click unsubscribe and List-Unsubscribe header. Non-compliant sends fail the quality gate.
• ·Strategy leakage from the sender's own site. Sends may only reference what is publicly visible on the sender's homepage and public product surface. Rejected drafts are logged, never delivered.

The pre-delivery quality gates

Every send passes all of these before delivery. Any single failure blocks delivery. No manual override.

The post-delivery audit

After delivery, each send is monitored via Google Postmaster Tools, Microsoft SNDS, and ESP feedback loops:

• ·Inbox placement verdicts pulled per domain and per batch within 2 to 48 hours.
• ·Verdicts of “delivered to spam” trigger a reputation-repair pass. The send's lineage records the event.
• ·Verdicts indicating sender-side issues (authentication failures, bounce surges, complaint spikes) pause the pipeline and notify the operator.
• ·User-side issues (recipient blocks, list-quality problems) are reported, never silently fixed.

Network safety, enforcement timeline

Domains join the network only after passing a trust check. Admitted domains are re-audited weekly. When a domain trips detection, this timeline runs publicly, with a case number.

What we publish, publicly, for accountability

Provider alignment map

Each row names a mailbox-provider guideline and the exact Outboxr mechanism that enforces it.

How to verify any of this

Pick any send from any domain in the network. Open its /send/:id/lineage page. You will see: the research sources used, the model and pipeline version, each quality gate's result, the link-integrity report, the consent check result, the delivery timestamp, the reputation band at send time, and any post-delivery heal events.

If anything in this section does not match what you see in a lineage page, email standards@outboxr.dev. Confirmed inconsistencies are corrected within 72 hours and logged in the standards history.

What this section does not promise

• ·We do not promise inbox placement. We promise we do not violate provider guidelines and that we publish the audit trail to prove it.
• ·We do not promise zero errors. We promise public accountability when we make them.
• ·We do not promise that every send will reach the inbox. We promise we respond to provider verdicts and iterate.

How sends composed through Outboxr are researched, written, gated, and disclosed.

Version v1.0 · Published 2026-04-18

This section governs content production for every send delivered through Outboxr, as an application of the broader Standards. Standards take precedence if conflicts arise.

Platform disclosure

Sends are researched and composed with AI assistance by the sender. The pipeline draws on the sender's own site, their recent public shipping, the recipient's public footprint (for outreach), and curated industry signals. Every send at scale carries a visible AI-assisted disclosure footer linking to this section.

What we guarantee

• ·Sends pass the pre-delivery quality gates listed in Standards before leaving the pipeline: minimum and maximum word bands, consent check, link integrity, SPF/DKIM/DMARC, banned-phrase filter, exposure-manifest compliance, reputation match.
• ·Sources are public. Research uses first-party news APIs, the sender's own site, and the recipient's public profile. Research sources are listed on each send's public lineage page.
• ·No paid placements in sender content. Outboxr does not insert advertising, sponsored sections, or affiliate links into any send.
• ·No behavioural tracking on the outboxr.dev frontend. No third-party fingerprinting. No data sold.
• ·Every bulk send includes an AI-assisted disclosure footer and a visible lineage audit trail link.

What we are not

• ·Not a spam farm. Every sending domain is publicly listed with inspectable reputation and consent-ledger counts.
• ·Not a scraper. Research uses first-party APIs and public sources without cloning or aggregating recipient data.
• ·Not a blast tool. Each send is uniquely composed for the domain and the segment. Reputation-band throttling is non-negotiable.
• ·Not opaque. Prompts, model versions, sources, validation results, delivery verdicts, and link-integrity reports are published per send.

Error correction process

1. ·Report errors to corrections@outboxr.dev with the send ID and error description.
2. ·Review occurs within 72 hours. Confirmed errors receive visible correction notices and lineage logging.
3. ·Systematic errors trigger rule changes documented in the standards history.

What may and may not be sent through Outboxr, enforcement procedures, and appeals.

Version v1.0 · Published 2026-04-18

Prohibited behavior

Severe violations (immediate permanent removal)

• ·Child sexual abuse material (CSAM). Any depiction triggers reporting to NCMEC, domain removal, permanent operator ban, and data preservation for law enforcement.
• ·Credible threats of violence. Content threatening specific people or groups with physical harm is reported to relevant authorities and results in immediate permanent removal.
• ·Doxxing and targeted harassment. Distributing private personal information to harass, intimidate, or enable targeted violence.
• ·Phishing and malware distribution. Credential harvesting, social engineering, fake transactional notifications, or links to malicious payloads. Never permitted under any circumstance.

High-severity violations (quarantine then permanent removal)

• ·Purchased, scraped, or unconsented lists. Sending to addresses without a verifiable consent record.
• ·Deceptive headers or spoofed From addresses. Misrepresenting the sender identity, forging thread history, or bypassing the From-alignment check.
• ·Broken or missing unsubscribe. One-click unsubscribe is mandatory on every bulk send. List-Unsubscribe headers must resolve.
• ·Spam-farm behavior. Coordinating send schedules, sender identities, or content across unrelated domains to launder reputation.
• ·Cloaked content. Sending one message to the ESP preview environment and a different message to real recipients.
• ·Hidden text and filter-evasion tricks. Zero-width characters, invisible text, homograph spoofing, base64 payloads engineered to defeat spam classifiers.
• ·Thin or filler content. Bulk sends consisting of empty boilerplate, scraped aggregations, or auto-generated stubs without editorial substance.
• ·Deceptive monetization. Undisclosed affiliate promotions, fake reviews, misrepresented product claims, fraudulent transactional-style solicitations.

Standard violations (quarantine with full review and appeal)

• ·Audience overlap. New domains with more than 70% recipient overlap with an existing member pause admission until differentiation occurs.
• ·Drift into prohibited categories. Admitted domains gradually sending inconsistent content trigger re-audit, not immediate removal.
• ·Persistent quality-gate failures. More than 30% of pipeline runs failing over 30 days indicates configuration or strategy problems. The pipeline pauses pending review.
• ·Abuse of the appeals process. Three bad-faith appeals within 90 days trigger manual review of all future appeals from that operator.

Enforcement timeline

Appeal process

1. ·Visit /appeal or email appeals@outboxr.dev with the case number in the subject line.
2. ·Describe why the flag is incorrect. Include screenshots, logs, or counter-evidence.
3. ·A human reviewer (not the automated system) evaluates within 48 hours.
4. ·Decision, reasoning, and rule adjustments are published in the transparency report with case number.

Successful appeals restore full access and remove spam-index entries within 2 hours.

Outboxr commitments

• ·Evidence before action. We do not enforce on rumour, competitive report, or vibes.
• ·Detection thresholds err toward not-flagging. Borderline cases receive human review before quarantine.
• ·No removal based on competitive complaints. Only signal-driven detection counts.
• ·All overturned flags published in the transparency report with rule updates.
• ·Outboxr-operated domains face identical detection standards and quarantine timelines.

Reporting violations

Email abuse@outboxr.dev with the send ID or domain and the specific policy clause violated, plus evidence (headers, screenshots, archives, reproduction steps).

Response timeline: acknowledgment within 24 hours, detection pass within 48 hours for credible reports.

Copyright and DMCA: email legal@outboxr.dev with full DMCA notice per 17 U.S.C. §512(c)(3).

Repeat infringers

Operators removed for confirmed abuse face permanent registration blocks across email addresses, payment instruments, and infrastructure fingerprints. Appeals follow the standard process.

What we collect, where it lives, how long, who sees it, and how to get rid of it.

Version v1.0 · Published 2026-04-18

Who we are

“Outboxr,” “we,” or “us” refers to the service operated at outboxr.dev by Heimlandr (Sweden). We are the data controller for information collected through the Outboxr service.

Privacy contact: privacy@outboxr.dev.

What we collect, and why

Data you give us when you register a domain

Sending domain, From address, optional SMTP or ESP credentials (stored encrypted and redacted from API responses), voice preferences, branding (colours, logo path, footer text), send cadence and preferences, optional webhook URLs for delivery events.

Why. To send email from your domain the way you configured it.

Data we gather from your domain's public surface

When you register or are re-audited, our crawler fetches your public homepage, sitemap, and a bounded number of linked pages. We extract meta tags, headings, structured data, visible text, brand signals (colours, logo), and authentication hints (published SPF, DKIM, DMARC records).

Why. To compose sends matching your voice and niche, detect spam or phishing patterns before admission, and enforce the exposure-manifest rule that prevents sends from leaking claims not already public on your own site.

Consent records per recipient

For any bulk send we require a consent record tied to the recipient address: timestamp, source (double opt-in form, order confirmation, direct reply, explicit import with proof), and the minted consent-ledger entry. Unsubscribes append a termination entry to the same ledger.

Why.To comply with CAN-SPAM, GDPR, CASL, and every mailbox provider's bulk-sender requirement, and to verify consent at the consent-gate stage before delivery.

OAuth tokens for Google Postmaster Tools (optional)

If you connect Google Postmaster Tools, we store an encrypted refresh token (AES-256-GCM at rest) scoped to the read-only postmaster permission. We never request write access. You can revoke at any time from your Google account permissions page. The token is then invalidated at Google and at our side.

Why. To pull inbox placement verdicts per batch so the pipeline can auto-heal reputation and so you can see real delivery data instead of send-side opens.

Operational records

We log authentication events (key mint, revoke, failed-auth IP, timestamp), pipeline runs, delivery events (accept, bounce, complaint), error traces, and access logs for the public API. Logs are rotated and access-controlled.

Why. Security, abuse detection, debugging, and the public-audit obligations in our Standards.

Sends you dispatch through Outboxr

Composed sends are dispatched through your connected sending infrastructure (your own SMTP, your own ESP, or hosted mode). A searchable send registry index (subject, send ID, domain, segment, delivery verdict, timestamp) is kept for lineage and deliverability analytics.

Why. To deliver sends, to maintain consent and reputation records, and to publish the lineage record per send.

What we do not collect

• ·No behavioural tracking on the outboxr.dev frontend.
• ·No third-party analytics (Google Analytics, Mixpanel, etc.).
• ·No fingerprinting, no session replay, no heatmaps.
• ·No advertising identifiers.
• ·No inbox contents. Outboxr does not read the recipient's inbox. Reply detection uses your own mailbox via the IMAP or API credentials you provide.
• ·No payment card numbers. Stripe handles payment processing. Outboxr stores only Stripe customer IDs.
• ·No logs of send contents sent through the LLM once the pipeline finishes composing.

Where your data lives

Outboxr runs on servers in Germany (Hetzner, Falkenstein and Nuremberg region). Our database and encrypted backups live there. Our jurisdiction is the EU/EEA (Sweden-based operator, German infrastructure). International transfers are covered below.

Third parties that may see your data

Outboxr sends limited, purpose-bound data to the following services. Each is a processor under GDPR terms. We do not sell your data. We do not share it with advertisers.

How long we keep data

• ·Domain registration data: until you delete the domain via the API.
• ·Crawl and audit data: 90 days rolling, then deleted.
• ·Consent-ledger entries: retained for the life of the recipient relationship plus 6 years, per common data-protection requirements.
• ·Auth events (key mint, revoke, failures): 90 days.
• ·Pipeline run history: 180 days.
• ·Send registry entries: until you delete them via the API. Exportable anytime.
• ·OAuth tokens (Postmaster): until you disconnect or Google revokes.
• ·Encrypted backups: rolling 7 days.

International transfers

Outboxr stores your data within the EU/EEA. Some processors above are based outside the EU (Anthropic, Google, Postmark/AWS, Stripe) and receive data under the EU's Standard Contractual Clauses and the processor's own certification (DPF where applicable). If you are an EU/EEA resident, you have the right to request the specific transfer mechanism used for any processor.

Your rights

• ·Access all data we hold about your domain via the API.
• ·Export sends, consent records, and delivery logs via npx outboxr export.
• ·Delete a domain and all associated data instantly via DELETE /api/domains/:id.
• ·Disconnect Postmaster Tools at any time. Token revoked client-side and server-side.
• ·Object to any specific processing by emailing privacy@outboxr.dev.

EU/EEA residents additionally have the right to lodge a complaint with their local Data Protection Authority. UK residents may contact the ICO. California residents have CCPA rights. Contact us and we respond within 45 days.

Cookies

The outboxr.dev frontend is a static site with no analytics and no advertising cookies. We may set a short-lived signed cookie during OAuth flows (10-minute TTL) solely to prevent CSRF. No third-party cookies are set.

Security

Tenant API keys are stored as SHA-256 hashes. We cannot recover them. Postmaster refresh tokens are encrypted at rest with AES-256-GCM. All traffic is HTTPS-only with HSTS enforced. Breaches that may affect user data are disclosed within 72 hours of discovery as required by GDPR Art. 33.

Children

Outboxr is for developers, marketers, and domain operators, not children. We do not knowingly collect data from anyone under 16. If you believe we have such data, email privacy@outboxr.dev and we will delete it.

Changes to this policy

Material changes are announced by email to active tenants at least 14 days before taking effect. Every prior version is archived at /privacy/history. The version identifier and publish date at the top of this section is the source of truth.

The legal baseline for using Outboxr.

Version v1.0 · Published 2026-04-18

1. Accepting these terms

By registering a domain with Outboxr, running npx outboxr init, or using the API, you agree to the Terms of Service, Privacy Policy, Acceptable Use Policy, and Standards. If you disagree, do not use the service.

If you use Outboxr on behalf of an organization, you confirm authority to bind that entity to these terms.

2. Who we are

Outboxr is operated by Heimlandr, registered in Sweden. “We,” “us,” and “Outboxr” refer to this legal entity. “You” means the person or organization using the service.

3. What Outboxr does

Outboxr composes newsletters and 1:1 outreach sends, routes them through reputation-aware sender identities in a trusted network, runs deliverability audits, tracks inbox placement, and optionally pulls Google Postmaster Tools signals to auto-heal reputation. Full features are described on this landing page and enforced per the Standards.

4. Eligibility

You must be at least 16 years old and able to enter binding contracts in your jurisdiction. You must not be on sanctions lists or in jurisdictions where providing the service is illegal.

5. Your account and API keys

5.1 Tenant keys

When registering a domain, a tenant API key scoped to that domain is created. The raw key displays once and is never shown again. Only a SHA-256 hash is stored. You are responsible for securing the raw key. Upon suspected exposure, revoke and mint a new key immediately.

5.2 Responsibility for actions

Any action performed with a valid key is attributed to the corresponding domain. All key events (mint, revoke, auth failure) are logged in the audit trail.

6. Fees and billing

6.1 Free tier

The free tier (weekly cadence, one sending domain, full quality gates, full consent ledger) requires no payment and has no expiry. Current features are listed on the pricing section.

6.2 Pro tier

Pro tier billing is monthly via Stripe at the per-domain price shown on pricing. Rates are degressive. Per-domain cost decreases with additional domains. Pricing at signup applies to each cycle. Changes affect future cycles only with 30 days' notice minimum.

6.3 Taxes

Displayed prices exclude VAT or applicable taxes, added at checkout where legally required.

6.4 Cancellation

Cancel anytime via npx outboxr disconnect. Pro features cease at the current billing period's end. No partial-period refunds except where consumer-protection law requires otherwise.

7. Acceptable use

What may and may not be sent through Outboxr, enforcement procedures, and appeal processes are detailed in the Acceptable Use Policy, which is part of this agreement. Violations may result in quarantine or permanent removal per the timelines described there.

8. Your content

8.1 You own it

Every send composed and dispatched through Outboxr belongs to you. Text, templates, metadata, and outputs are yours. Outboxr claims no ownership.

8.2 The license you grant us

To provide the service, you grant Outboxr a worldwide, non-exclusive, royalty-free license to: (a) process content through the pipeline to compose, gate, and dispatch sends, (b) display excerpts and metadata in the public send registry and delivery-audit reports, (c) include send lineage in transparency reports, and (d) retain consent records per retention schedule. This license ends upon content deletion or domain termination, except where retention is legally required.

8.3 Responsibility for content

You ensure dispatched content, including Outboxr-composed content you approve, complies with applicable law, does not infringe third-party rights, holds valid consent from every recipient, and does not violate the Acceptable Use Policy.

8.4 AI-generated content

Outboxr composes sends using large language models. Quality gates per Standards are applied, but factual correctness, originality, or legal safety for every jurisdiction cannot be guaranteed. As sender, review outputs before high-stakes use.

9. Our intellectual property

Outboxr software, brand, outboxr.dev content, Standards, and related IP are owned by Heimlandr and licensed for service use only. Copying, reselling, reverse engineering, or rebranding without written permission is prohibited. The Outboxr engine may be released under an open-source license. Releases are documented in the standards history.

10. Service availability

High availability is aimed for but no specific uptime guarantee exists for free or Pro tiers. Scheduled maintenance is communicated via the status channel. Enterprise SLA terms, when available, are offered in separate written agreements.

11. Disclaimers

The service is provided “as is” and “as available.” To the fullest permitted extent, all warranties, express or implied, including fitness for particular purpose, merchantability, and non-infringement, are disclaimed. No warranty covers specific inbox-placement rates, open rates, reply rates, or business outcomes.

Consumer rights under EU/EEA or equivalent law are not waived. Where legally non-disclaimable, warranties apply only to legally required minimums.

12. Limitation of liability

To the fullest permitted extent:

• ·Indirect, incidental, special, consequential, or punitive damages are excluded, including loss of profit, revenue, deliverability, or data.
• ·Total aggregate liability for any claim is capped at the greater of (a) amounts paid in the preceding 12 months, or (b) €100.

These limits apply regardless of legal theory (contract, tort, statute) and whether damages were foreseeable. Where certain limitations are prohibited, they apply maximally within local law.

13. Indemnity

You defend and indemnify Heimlandr against third-party claims from (a) your content, (b) service use violating these Terms or the Acceptable Use Policy, or (c) applicable law violations including missing consent or recipient-protection rules. Heimlandr notifies you promptly of covered claims and cooperates in defense.

14. Termination

14.1 By you

Run npx outboxr disconnect or call DELETE /api/domains/:id. Data is deleted per the Privacy Policy retention schedule.

14.2 By us

Immediate suspension or termination occurs for serious or repeated Acceptable Use Policy violations, payment failure, or legal compulsion. Non-urgent terminations include 30 days' notice. On termination, sections 8, 9, 11, 12, 13, and 15 survive.

15. Governing law and disputes

These Terms are governed by Swedish law. Disputes are resolved in Stockholm courts unless local consumer-protection law requires otherwise. Nothing waives statutory consumer rights.

16. Changes to these terms

Material changes are emailed to active tenants at least 14 days before taking effect. Continued use after the effective date constitutes acceptance. All prior versions are archived. The version identifier and date at the top of this section is authoritative.

17. Miscellaneous

If any part is held unenforceable, the remainder stays in effect. Failure to enforce a right does not waive it. Assignment requires consent. Heimlandr may assign to affiliates or successors. No third-party beneficiaries exist.

18. Contact

Legal questions, service notices, DMCA claims: legal@outboxr.dev.