NewRAGSuite Engine: the knowledge engine for your data, run in Germany.About RAGSuite Engine

ProductRAG enginePlatform overviewAI governanceAI chatbotBuild AI assistantsAI searchAI knowledge baseConnectors & integrationsMCP serverTake the product tour →
Engine & self-hostedRAGSuite EngineSelf-hosted / on-premiseOpen source (Community)RAGSuite Engine or self-hosted? →
SolutionsSMEsHealthcarePublic sectorBanking & insuranceChatGPT alternativeAgencies & partnersAll solutions →
Case studies
Pricing
ResourcesGuidesBlogFree toolsAI data protectionEU AI ActTrust Center
Open source on GitHub
Start free trialBook a demo
Platform · AI governance

AI governance: control and evidence for every AI answer.

With RAGSuite you decide who sees which knowledge, which language model answers and how long data is kept. Every action is in the audit log. This covers AI Search, AI Assistant and AI Connectors, in RAGSuite Engine in Germany and in Enterprise on your own servers. So IT, data protection and the works council can approve AI.

Governance · HR projectExample
Access
TeamHR · 3 roles
SourcesHR handbook, policies
Sign-in
Sign in with Google
Two-factor authentication
Data and model
Language modellocal
Chat history90 days
Audit logfull · export
The problem

Without governance, AI stays a pilot.

Staff use private AI accounts with company data, nobody knows where data is processed, and when the works council or internal audit asks, there is no evidence. RAGSuite turns those questions into settings in the platform.

AI without governance
  • Private accounts with company data
  • Everyone sees everything, or nothing
  • No record of who asked what
  • Unclear how long data is kept
RAGSuite
  • Roles, teams and project access
  • Each team sees only its own sources
  • Full audit log with export
  • Deletion periods with receipts
Building blocks

What you control in RAGSuite.

Access

Organisations, teams, roles and project access. Sign-in with Google, two-factor sign-in, session timeout.

Audit log

Event type, user, project, action and severity. Complete, filterable, exportable.

Retention

Storage can be switched off, deletion periods with receipts, rules per source, compliance exports.

Models and quality

Model choice per project, model comparison and query tracing. A source on every answer.

Access

Roles, teams and sign-in: who may do what?

The Community Edition has two roles, admin and member, and sign-in with email and password. In RAGSuite Engine and in Enterprise you set up organisation, teams and roles and share projects selectively. Google sign-in is added on top.

Access rights from source systems such as SharePoint are not carried over automatically; you separate content by project.

Team & rolesEnterprise
AWAnna WeberAdmin
JSJonas SchulzEditor
MKMehmet KayaUser
LHLaura HoffmannUser
Sign-in & security
Sign in with Google
Two-factor authentication
Session timeout30 minutes
Accountability

The audit log: who did what, and when?

The audit log records every action with event type, user, project, action and severity. The Community Edition shows the last 15 days. In RAGSuite Engine and in Enterprise the log is complete, can be filtered and exported for internal audit, data protection and supervisors.

Audit logEnterprise
All actionsLast 30 daysExport
TimeUserAction
24 Sep 09:14a.weberSource added
24 Sep 08:52a.weberRole changed
23 Sep 17:30p.neumannModel switched
23 Sep 11:05a.weberRetention changed
22 Sep 16:48l.hoffmannSign-in
22 Sep 10:12a.weberExport created
Retention

Store, delete, prove.

Conversation storage can be switched off for chat and search separately. Automatic deletion periods remove old history, with receipts. In RAGSuite Engine and in Enterprise, retention rules per source and compliance exports are added.

AI data protection →

Data retentionSettings
Storage
Store conversations
Store search queries
Deletion periods
Chat history90 days
Analytics data30 days
Recent deletion runs
24 Sep 2026 · 03:00Receipt
23 Sep 2026 · 03:00Receipt
Models and data flow

Where does your data go? You decide.

In RAGSuite Engine, documents, search index and history live in a data centre in Germany. Self-hosted, everything stays on your servers, with a local model, nothing leaves your network at all. If you use an external language model, only the question and the matching passages go to that provider. You set this per project. Here is what that looks like self-hosted:

Always on your servers
DocumentsSearch indexHistoryUser accounts
+ language model: local, e.g. via Ollama
Local model (e.g. via Ollama)

Nothing leaves the building. Works with no internet connection (air-gapped).

Data leaving: none

RAGSuite itself never sends data to us. No phone-home, no telemetry. Compare models →

Jürgen Pietschmann
Not sure which language model fits your data?Jürgen Pietschmann · AI Consultant · RAGSuite Lead
Set-up

Set up governance in four steps.

In RAGSuite, projects are the boundary for knowledge. Each project has its own sources, settings and widgets. That ensures a team only gets answers from content it is allowed to see.

01

Define the cut

One project per team or confidentiality level, such as “All staff”, “HR”, “Management”.

02

Assign sources

Only the libraries, spaces and files that all users of the project may see.

03

Grant access

Assign teams and roles to the project, in RAGSuite Engine and Enterprise.

04

Review

Ask test questions and check in the audit log who accesses what.

Sign-off

What each team needs to approve AI.

IT

Roles, sign-in with Google, audit log, running on your own servers and no phone-home.

Data protection

DPA, DPIA template, deletion periods and data location in Germany.

Works council

Clear rules on who sees what, and an AI usage policy template.

Internal audit

Audit export, a source on every answer and the Trust Center documents.

Plans

What is in which plan?

FeatureCommunityRAGSuite EngineEnterprise
2FA, sessions, sign-in notifications✓✓✓
System status✓✓ (run by RAGSuite)✓
Rolesadmin and memberroles, teams, project accessroles, teams, project access
Sign-inemail and password+ Google+ Google
Audit loglast 15 daysfull, filters, exportfull, filters, export
Storage off, deletion periods✓✓✓
Retention per source, compliance exports–✓✓
Model comparison, query tracing, advanced analytics–✓✓
HostingYouRAGSuite, data centre in GermanyYou or a dedicated instance
White-label, air-gapped––by agreement
SupportCommunityEmail DE/ENEmail DE/EN, SLA by agreement

All pricing →

FAQ

Questions about AI governance.

What does AI governance mean?

AI governance is the set of rules and evidence a company uses to run AI under control: who has access, which data is used, which language model answers, and how that is documented. In RAGSuite, these rules are settings in the platform.

What governance is in the free Community Edition?

The basics: two-factor sign-in, session management, admin and member roles, storage that can be switched off, deletion periods and a 15-day audit log. Organisations, teams and project access, sign-in with Google, the full audit log and retention rules per source come with RAGSuite Engine and Enterprise.

Does RAGSuite support single sign-on?

In RAGSuite Engine and Enterprise, users sign in with their Google account. We only name further directory connections once they are available. Ask us about the current status for your environment.

Does RAGSuite help with the EU AI Act?

RAGSuite provides building blocks for it: the audit log, a source on every answer and control over which model answers. The duties under the AI Act stay with your company. Our EU AI Act guide shows what applies now.

Are permissions from SharePoint or Confluence carried over?

Not automatically, today. Separate content by project: each project has its own sources and settings. In RAGSuite Engine and Enterprise you also decide which teams can access which project.

Stefan Reinhardt
Stefan Reinhardt
Public Sector & Compliance

Show your review team the governance live.

We show roles, the audit log and deletion periods on a setup like yours, or you test it yourself in RAGSuite Engine.