Terms of Service
Version 1.0. Effective 2026-10-20.
1. Who we are and what these terms cover
loremem ("we", "us") is a hosted memory service for AI agents. It is run by an individual developer. It is not run by a company. You can reach us at l1106842466@gmail.com.
These terms cover:
- the Console at app.loremem.com;
- the API at api.loremem.com;
- the plugins and command-line tools when you use them with the hosted service.
The open-source Lore software that you run yourself is covered by its own license. It is not covered by these terms.
You accept these terms by ticking the agreement box when you sign in. If you do not agree, do not use the service.
2. Accounts
You must be at least 16 years old to use the service.
You may sign in with an email and password, or with a third-party account such as GitHub, Google or LINUX DO. You are responsible for your account and for every API token you create. You are responsible for anything done with them, including actions taken by AI agents you connect.
If you think a token or your account has been compromised, revoke the token in the Console and tell us.
3. Acceptable use
Do not use the service to:
- break the law or infringe anyone's rights;
- store or distribute malware, or content you have no right to store;
- attack, probe or overload the service, or get around its limits or access controls;
- access another user's data;
- resell the service without our written permission.
We may suspend or remove content or accounts that break these rules.
4. Your content
Your content is everything you or your agents store in the service: memories, skills, project settings and secrets. It stays yours.
You give us permission to store, process and transmit your content only as needed to run the service for you. This includes:
- sending text to the third-party AI model providers we use for search, recall and Dream (background memory maintenance);
- keeping backups.
We do not use your content to train models, and we do not sell it.
You are responsible for having the right to store your content. You are also responsible for how your agents use it.
5. AI processing and automated changes
Recall results, Dream changes and other model-generated output can be wrong, incomplete or out of date.
Dream may rewrite, merge, move or delete memories on its own. Each change is recorded so you can review or roll it back, but check anything important before you rely on it.
6. Early-stage service
The service is in an early stage and is provided free of charge. Features may change, be limited or be removed at any time. Data loss is possible. Keep your own copies of anything important.
The service is provided "as is" and "as available", without warranties of any kind, to the extent the law allows.
7. Paid plans
The service is currently free. If we introduce paid plans, the price and any additional terms will be shown to you before you buy. Paying is always your choice. Existing free features will not start charging you without notice.
8. Limitation of liability
To the extent the law allows, we are not liable for any indirect, incidental, special or consequential loss. This includes lost profits, lost data and business interruption.
Our total liability for any claim relating to the service is limited to the greater of:
- the amount you paid us in the 12 months before the claim; and
- 50 US dollars.
Nothing in these terms limits liability that cannot be limited by law, or takes away rights you have as a consumer under the law of the place you live.
9. Suspension and termination
You may stop using the service at any time. You can ask us to delete your account and data by writing to l1106842466@gmail.com.
We may suspend or close an account that breaks these terms or puts the service or other users at risk. If we decide to shut the service down, we will give at least 30 days' notice where we reasonably can, so you have time to export your data.
10. Changes to these terms
Each version of these terms has a number.
- Minor versions (for example 1.0 to 1.1) fix wording or add clarifications. They take effect when published.
- Major versions (for example 1.x to 2.0) change your rights or obligations. We will ask you to accept the new version before you continue using the service.
We keep a record of which version you accepted and when.
11. Governing law
These terms are governed by the laws of the Hong Kong Special Administrative Region. The courts of Hong Kong have non-exclusive jurisdiction over any dispute.
This does not take away any mandatory consumer protection that applies where you live.
12. Feature-specific terms
Some features carry extra risk. When you turn one of them on, you are asked to accept its section below, and the acceptance is recorded. Where a section conflicts with the rest of these terms for that feature, the section applies.
12.1 Secrets
Secrets lets you store credentials, such as API keys and tokens, so your agents can use them by name.
How secrets are stored
- Secrets are encrypted on our servers.
- The encryption key is held by us as the operator of the service.
- The Console never shows a stored value again.
Who can read a secret
- Any program that holds one of your Lore API tokens can read any of your secrets. This includes every AI agent on any machine where the token is configured.
- Lore cannot tell your agents apart from each other, or from other software using the same token.
Where values can end up
- A value that is read is placed into the command that uses it.
- If that command prints the value, it can end up in conversation logs and with the AI model provider you use.
What you are responsible for
- What you store.
- Using the narrowest, revocable, time-limited credentials you can.
- Rotating any credential you think was exposed.
Do not store high-value credentials, such as banking credentials, cloud root or owner accounts, or production root passwords.
What we record
- Every read is logged with the API token used and the time.
- You can delete a secret at any time.
- If you turn the feature off, stored secrets stay but cannot be read until you turn it on again.
We are not liable for losses caused by a stored credential being read or used by software you authorised through your API tokens, to the extent the law allows.