Terms of Service
Last updated: 12 November 2026
1. The service
backendup ("the Service") is an online collaborative work tracker provided by the operator of this domain ("we", "us"). By creating an account or using the Service you agree to these terms. If you use the Service on behalf of an organization, you confirm that you may bind that organization.
2. Accounts
You must provide a valid email address, keep your credentials confidential, and be honest about your identity. You are responsible for activity on your account. Notify us promptly of suspected misuse. Accounts may be signed in via email and password or via a supported third-party provider; the provider's own terms apply to that sign-in step.
3. Acceptable use
You agree not to: misuse or attempt to gain unauthorized access to the Service or other accounts; upload unlawful, malicious, or infringing content; spam or resell access without our consent; overload the Service (for example by abusive API use), or use it to violate applicable law. We may suspend or terminate accounts that breach these terms, and we will explain the reason when reasonable to do so.
4. Your content
You (or your organization) keep ownership of the content you submit. You grant us only the narrow licence needed to store, display, transmit and back up that content to provide the Service, and to show it to the people you share it with. We do not train AI models on your private work content.
5. Availability & changes
The Service is provided "as is", without a guaranteed uptime. We may change, update or discontinue features; for material reductions we will try to give reasonable notice. Exact availability commitments may be added to paid agreements.
6. Liability
To the extent permitted by law, we are not liable for indirect or consequential losses, and our total liability for claims arising from your use of the Service is limited to the amount you paid for it in the preceding twelve months (or €100 for free accounts). Nothing in these terms limits liability that cannot be limited by law.
7. Plans, limits and fair use
Some plans carry published usage limits. The numbers in force are the ones rendered live on our usage-limits page - they are read from the same settings the server enforces, and we never publish one number and enforce another. Limits exist to keep a shared service affordable and abuse-proof; they bound cost, they do not meter usefulness. Work items, messages and history are not limited on any plan, ever.
The promise that holds on every plan, free or contracted, stated verbatim: a limit can pause a new action; it never deletes, hides or locks anything you already have. Reaching a cap is "not right now", never a "goodbye".
8. Enterprise & Dedicated
Under a separate agreement we may raise or remove published limits for your organization tree, within the fair-use envelope set out in Appendix A - contracted generosity is a plan row, not physics. This tier is also where the operational commitments live, priced as what they cost rather than sold as quota: single sign-on (SAML), audit logs, data residency, retention controls, dedicated inference capacity, seat autoscaling as your roster grows, and support terms. None of these are ever a free-plan feature wall.
Self-hosting remains the unbounded answer: every limit value in this product is deployment configuration. Run the software on your own infrastructure and there is no envelope at all - only the terms of its own licence and your own machines.
9. Termination
You may stop using the Service and close your account at any time via your profile or by contacting support. On termination your access ends and content is removed according to the retention rules in the Privacy Policy.
10. Governing law & contact
These terms are governed by the laws of Germany, excluding its conflict-of-law rules, without prejudice to mandatory consumer-protection law of your country of residence. Questions: [email protected]. See also our Privacy Policy.
Appendix A — The fair-use envelope
This appendix exists because we would rather publish our rails than hide them. Where a contract says "without limits", it means: no published plan limits. The absolute rails below still exist, they are deliberately far above anything honest work produces, and they are sized to stop abuse - floods, scripted signup farms, storage hoarding - not to meter customers. We do not enforce them against ordinary use; if one is ever reached we contact the organization's admin and pause the affected new actions, because the promise in section 7 is contractual here too: a limit can pause a new action; it never deletes, hides or locks anything you already have.
- API traffic: no single tree sends more than 10,000 requests per minute across all of its credentials, and no single credential more than 5,000 per minute.
- Mail: outbound notification and invitation mail is bounded at 50,000 per tree per day (recipient domains rate-limit us; this is about deliverability, not features).
- Files: no single stored file above 1 GB, and no tree above 100 TB without a stated arrangement in the contract itself.
- Flood guards: human anti-spam cooldowns (chat posting, failed-login throttles) apply identically on every plan - a paid plan never buys a bigger flood.
- Shared inference: the per-message guardrails published alongside the AI assistant limits bound one generation on shared compute; dedicated and bring-your-own endpoints are outside the envelope entirely - they are your machines.
The envelope's numbers may only ever be raised by us, never lowered below the published plan rows, and every change carries reasonable notice. The usage-limits page stays the live source for the numbers that actually apply.