Data sources

A data source gets its name and permissions in MeinGPT — that is where who may find it is decided. Folders on an Outpost machine are created there and only adopted here; how much a folder releases is decided by the Outpost too. Where its documents live is decided afterwards — and for some sources the answer is "on a machine at your site", which is exactly what the Outpost is for.

Who fetches the files?

How an Outpost source comes about

Two halves in two places, in this order:

1. In the Outpost: share the folder — pick the path and decide how much it releases.
2. In MeinGPT: adopt the reported folder under Shares and name it. That is where who may find it is decided.

There is no button in the Outpost that creates a data source. If you are looking for one, step 1 is missing.

For folders, the dialog in the Outpost knows exactly one source type: a folder path on this machine — a permanently mounted SMB share counts too. When the services preview is enabled, a database, a local MCP server, or an internal HTTP API is added through the same entry point: Share in the Outpost, then pick the matching type and set connection details and access rights. Unlike a folder, a reported service is not first adopted and named in MeinGPT — you decide directly who may use it. Details on status and access rights are under Console & operations → Database, MCP and HTTP. S3 storage, WebDAV and web crawler are connected exclusively in MeinGPT directly. If such a service is not reachable from the internet — an internal MinIO or a Nextcloud, say — that needs its own connection method (IP allowlisting, Enterprise Connection Network or VPN) instead of an Outpost — see Guide: On-premise.

The sources in detail

Needs an Outpost

Directly in MeinGPT, no Outpost needed

Connecting: check first, then save

• For S3, WebDAV and Confluence, MeinGPT checks the credentials before the source is created. A wrong password, a wrong bucket or an unreachable server is reported right in the dialog, together with a preview of the first entries.
• If the same folder, the same library or a parent location is already connected as a data source, the dialog points it out, so the same files are not copied twice.
• Reconnect opens the form with the saved settings; only the secret has to be entered again. Syncing starts right after.
• A source can be renamed at any time directly on its page.

What is the same for every source

1. The Outpost fetches the files into its working area on this machine.
2. It extracts text, splits it into passages, encodes them and stores them in the search index.
3. Later runs only process what changed.
4. What it cannot read, it skips — and records it by file extension in the folder's row under Shares.

Point 4 is the one to keep an eye on: a skipped file cannot be found, and the search tells nobody. → Console & operations

One exception: if a file was readable before and only its newest version is damaged, encrypted or empty, the last readable version stays findable until a readable one is saved. The console and MeinGPT list such files as a failed update, with the reason. Deleting or excluding the file also removes the old version from search.

Exception: for local NTFS folders, there is a beta feature, Windows user permissions (Beta), that honors each user's real NTFS permissions — see Who sees what for details.