1. Who we are
“Anvric” (“we”, “us”, “our”) provides the Anvric desktop application and related services. The legal entity and registered address operating Anvric will be named here on publication. For privacy questions, contact privacy@anvric.com.
2. What lives on your device
Your Context Vault — the documents, notes, embeddings, and entity graph that make up your fused memory — is built and stored locally on your device in an on-device database. By default it is not uploaded to us in full.
At-rest encryption of the local vault is optional and is not enabled by default. When it is off, your local vault is stored unencrypted on your machine and is protected by your operating system account. We will tell you clearly in the app when at-rest encryption is available and whether it is on.
3. What is sent off your device, and why
Anvric is not a fully offline product. The following data leaves your device for specific features. We do not claim that “your data never leaves your device”.
- Your vault content, by default. Unless you mark it otherwise, content you add is stored on your device andmirrored to your cloud vault. You can mark any source, or any kind of content, device-only — that content is then not mirrored. Mirroring is what happens if you make no choice.
- AI processing. When you run a query that uses an AI model, the relevant context and your prompt are sent to the AI model provider that serves the request, over an encrypted connection, so the model can generate an answer.
- Account and sync. Account information and data needed for optional cross-device sync and for our services are stored in our cloud (hosted on managed infrastructure). Transport is encrypted in transit (TLS).
- Publishing a Slice. If you choose to publish a Slice to the Anvric Garden, the content of that Slice is uploaded to our cloud so that authorized buyers and their AI tools can query it. You choose what goes into a Slice; nothing is published without your action.
- Product analytics. We collect limited, privacy-minded usage and diagnostic data to operate and improve the product. You can ask us what we hold about you.
4. How we protect data we hold
- Data in transit is encrypted using TLS.
- Cloud data is isolated per account using row-level security, so one account cannot read another account's data.
- Before a Slice is published, we screen its content for detected secrets and personal information (using pattern-based and model-based checks) and block publication when a credential is detected. No automated screen is perfect — review what you publish.
We do not currently hold any HIPAA, SOC 2, ISO 27001, or PCI certification, and we do not claim one. We will only state a certification once it has actually been completed.
5. AI providers and training
We do not train our own foundation models on your private vault content. Queries you run may be processed by third-party AI providers; their handling of data, including any retention, is governed by their own data policies, which vary by provider. Where a provider offers a no-retention option, we use it where applicable.
6. Your controls and rights
- You can delete documents from your vault at any time.
- You can unpublish a Slice, which removes it from the Garden.
- You can request access to, correction of, export of, or deletion of the personal data we hold about you by emailing privacy@anvric.com.
Depending on where you live, you may have rights under laws such as the GDPR or CCPA. We aim to honour those rights for all users. Stating that we aim to honour these rights is not the same as a certification of compliance.
7. Children
Anvric is not directed to children under 16, and we do not knowingly collect their personal data.
8. Changes
We may update this policy as the product changes. When we make a material change we will update the effective date and, where appropriate, notify you in the app.