Trust & Security

Security you can verify

Security isn't a checkbox for us — it's the point. Pragor exists to make AI work accountable and durable, so trust is built into the product: strict tenant isolation, a complete audit trail, encrypted secrets, and full control over your data — including running Pragor entirely on your own infrastructure. We never use your data to train any model.

Strict tenant isolation

Every customer sees only their own projects — enforced server-side, not in the UI. Internal projects are invisible to tenants.

Complete audit trail

Every message, task, approval and operation is logged and attributable — who did what, when and why.

Encrypted secrets

Agent and integration credentials are encrypted at rest; connection passwords are hashed and rotatable.

We never train on your data

Your content is used only to run the service. It is never used to train any AI model — full stop.

Your data, your control

Export or delete your data at any time (GDPR). EU data residency is available.

Self-host / on-prem

Run Pragor entirely on your own infrastructure — air-gap friendly. Your servers, your data.

Human approval gates

Risky, irreversible actions can be gated behind an explicit human decision — nothing runs unattended.

Roles & least privilege

Operators, PMs and workers have clear, scoped authority; only authorised roles can take sensitive actions.

Transparent usage

Every billable event is accounted for and visible in your usage meters — no surprises.

Compliance roadmap

We practise the controls behind formal certifications today; audited certifications (e.g. SOC 2) are on our roadmap. If you need a security review or a signed DPA for a Business/Enterprise deployment, we'll work with you.

Request a security review / DPA Read the privacy policy