Security and Compliance

What the Outpost discloses, what stays on your machine, and what we can do

The Outpost runs on your network and indexes files that never leave your machine. What meinGPT gets to see in order to answer a question is your decision — and that setting is what this page is about.

Attention

On-premise does not mean "nothing leaves". It means: only what you release. An answer without the passage it rests on would be no answer, so at minimum the matching passage travels to meinGPT with every question. How much more is up to you.

The release level — the one setting that matters

Every data pool has a release level. It decides what leaves the machine when someone asks a question.

LevelWhat travels to meinGPT
PassagesThe matching passages and the file name. Answers get choppy — the model never sees a document's context, only fragments.
Full textThe complete text of the relevant documents. The default.
Original filesThe file itself. Needed when people should verify an answer against the original, or reuse a template.

There is deliberately no level below passages: a citation nobody can place is worthless.

An administrator sets the level in the web console — per pool, or as the default for the whole Outpost. The Outpost cannot cap that today; it does what the platform asks. A local ceiling ("never release more than full text from this machine") is planned and not yet built. It is stated here because it is the first thing a security review finds.

The connection

Outbound only

The Outpost opens the connection itself. No inbound port, no firewall rule, no service exposed to the internet.

A key per organization

The tunnel is signed with a key that belongs to your organization alone — not with a shared secret.

Technically it is an outbound, TLS-encrypted WebSocket connection to the meinGPT Bridge — not a VPN. Your network team has nothing to open; outbound HTTPS is enough. Corporate proxies and private root certificates are supported, and the setup assistant checks them.

What stays on the machine

The platform cannot set any of these. They exist only on the device, by design: if the cloud could add a folder, the cloud could read anything on that machine.

  • Which folders are shared. This is the actual boundary.
  • Which local services are reachable, and with which credentials. The password is never transmitted to meinGPT.
  • The root certificate for your TLS inspection.
  • Where embeddings go — the address of the model that turns your documents into vectors.

What we can do

The Outpost updates itself from storage we operate, signed with a key we hold. That carries an uncomfortable but honest limitation: we could ship a version that behaves differently from the one described here. Everything on this page is, today, a commitment rather than a technical proof.

To close that gap, turn auto-update off and pin a version you have reviewed. The price is that you apply updates yourself — in exchange, nothing rests on our restraint any more.

Revoking access

Both directions, with the same effect:

  • In the web console ("Unpair"): the secret is rotated and outstanding pairing codes are deleted. The right move when a machine is stolen, decommissioned, or no longer reachable.
  • On the device ("Disconnect"): the same — the Outpost revokes itself at meinGPT before clearing anything locally. If meinGPT cannot be reached, the app says so rather than reporting success.

Both are recorded in your organization's audit log.

Further reading

Note

Outpost-specific privacy: more on the architecture and how your data is handled is in the Outpost privacy documentation.

Was this page helpful?