Trust Center

Your data, your users and your agents, protected

How DNotifier keeps your data private and your AI agents under control. If your security team needs something that isn't on this page, email us and we'll help.

At a glance

The commitments that apply to every DNotifier customer.

We don't train on your data

Your prompts, documents, messages and outputs are never used to train AI models.

You own your data

Everything you store in DNotifier belongs to you.

Encrypted in transit

HTTPS for the API and dashboard. Secure WebSockets (wss://) for realtime.

Encrypted at rest

Stored messages, chat history and knowledge base documents are encrypted.

SSO on every plan

Microsoft Entra ID, Google, or your own identity provider over OIDC or SAML 2.0. Free plan included.

Roles and audit logs

Invite your team with roles. Sign-ins and account actions are logged with time, user and IP address.

Agents under control

Per-tool permissions, human approval steps and a full trace of every agent run.

Your region on Enterprise

Enterprise plans are deployed in the cloud region you choose.

Your data

DNotifier stores what your application needs to run, and nothing else is done with it.

Our commitment: we do not use your prompts, documents, knowledge base content, messages, chat history or AI outputs to train AI models, ours or anyone else's.
  • You own your dataMessages, chat history, knowledge base documents and agent outputs belong to you.
  • Your choice of model providerWhen an agent calls a model, the request goes to the provider you enabled for that agent, such as OpenAI, Anthropic, Google, AWS Bedrock or Azure AI, under that provider's API terms. Check each provider's data policy for your use case.
  • Run models on your own hardwareConnect Ollama to keep model calls inside your own infrastructure.
  • Retention that fits your policiesEnterprise plans get message retention and event handling tailored to your requirements.

Encryption

Data is protected on the way in, on the way out and while it is stored.

  • In transit: HTTPSAll traffic to the DNotifier API, the dashboard at app.dnotifier.com and the knowledge base REST API uses HTTPS (TLS).
  • In transit: secure WebSocketsRealtime messaging and agent events run over encrypted WebSocket connections (wss://).
  • At restStored data, including messages, chat history and knowledge base documents, is encrypted at rest.

Access control

Decide who can reach your projects and see what they did.

Single sign-on is included on every plan, including Free. Many platforms keep SSO for their enterprise tier.
  • Single sign-onMicrosoft Entra ID, Google, or your own identity provider over OIDC or SAML 2.0. Test the connection before you switch it on for your organisation.
  • Organisations, members and rolesInvite teammates to your organisation and manage their role and status from Settings → Members.
  • Audit logsSign-ins and account actions are recorded with the time, the user, the IP address and details, under Settings → Audit Logs.
  • Credentials you controlEach app has its own app ID and secret. Knowledge base REST API keys are generated per project and can be switched off. Keep secrets on your server, never in client code.

AI agent controls

Agents can only do what you allow, and you can see everything they did.

  • Per-tool permissionsEach agent lists the tools it may call. With permissionEnforcement: "backend", permissions are checked on the server, not only in your code.
  • Human approval stepsPut a person in the loop before an agent takes a sensitive action, such as a refund or sending an email.
  • Full run tracesWith observability: true, every agent and workflow run is recorded step by step: model, tools, inputs, outputs and timing.
  • MCP servers on your termsConnect MCP servers over HTTP with your own headers and see each connected tool and its permissions in the Agent Registry.

Report a security issue

If you think you've found a vulnerability, tell us privately.

  • Email support@dnotifier.comUse the subject line "Security". Include what you found, the steps to reproduce it and any affected URLs or app IDs.
  • Keep it private until it's fixedPlease don't share the issue publicly or access other customers' data while testing.
  • We'll keep you updatedWe'll acknowledge your report and tell you when it's resolved.

Report an issue

Documents for your review

Running a vendor assessment? We'll answer your security questionnaire and walk your team through the architecture.

FAQs

Frequently asked questions

Short answers to what security and procurement teams ask most.

Does DNotifier train AI models on my data?

No. DNotifier does not use your prompts, documents, knowledge base content, messages or AI outputs to train AI models. Your data is used only to run your application.

You do. Messages, chat history, knowledge base documents and agent outputs belong to you.

Yes. Traffic to the DNotifier API and dashboard uses HTTPS, realtime connections use secure WebSockets (wss://), and stored data is encrypted at rest.

SSO is included on every plan, including Free. You can sign in with Microsoft Entra ID or Google, or connect your own identity provider over OIDC or SAML 2.0.

Yes, on Enterprise plans. DNotifier is deployed in the cloud region of your choice.

Yes. You choose the model provider for each agent, and you can connect Ollama to run open models on your own hardware.

Email support@dnotifier.com with the subject "Security". Please include steps to reproduce and don't share the issue publicly until we have fixed it.

Email support@dnotifier.com. Tell us your use case for a demo, or the documents and questionnaire your security team needs.

See it with your own data

Book a demo and we'll show you SSO, audit logs, agent permissions and run traces on a project like yours.