Skip to Content
GuidesOverview

Guides

Use these guides to deploy, configure, and operate Eneo. Start with the task you need to finish; architecture pages are linked where the underlying design matters.

Start here

If you need to…Start with
Deploy EneoDeploy Eneo
Upgrade to 2.2.0 (renamed variables)Upgrading to 2.2.0
Upgrade to 1.7.0 (tenant models)Upgrading to 1.7.0
Upgrade legacy File/Icon storageUpgrade File & Icon Storage
Choose where durable bytes liveChoose Content Storage
Configure login and accessAuthentication & OIDC
Give each tenant its own IdPMulti-Tenant OIDC Federation
Provision users and groups from your IdPSCIM Provisioning
Connect an AI providerAI Provider Configuration
Enable web search or image generationWeb Search & Image Generation
Add documents to a knowledge baseDocument Processing
Connect SharePoint or OneDriveSharePoint Integration

All guides

Authentication & OIDC

Learn how to configure single sign-on using OpenID Connect (OIDC) with various identity providers:

  • Azure Entra ID (formerly Azure AD)
  • Keycloak
  • Auth0
  • Other OIDC-compliant providers

OIDC Federation

Manage identity providers through the federation API instead of environment variables:

  • Multi-Tenant OIDC Federation — each tenant authenticates against its own IdP; covers prerequisites, DNS, IdP registration, tenant configuration via API, tenant resolution, and troubleshooting
  • Single-Tenant Federation — one IdP for all users, managed through the federation API

SCIM Provisioning

Automate user and group provisioning from your identity provider with SCIM 2.0:

  • Generate, check, and revoke a tenant SCIM token
  • Configure Azure Entra ID provisioning and attribute mappings
  • Verify the integration and the email claim OIDC sign-in depends on
  • Understand provisioning scope, deprovisioning, and multi-tenant behavior

Audit Logging

Set up audit logging for compliance and security monitoring:

  • Enable audit logs and configure action toggles
  • Review access justification flow
  • Retention policy and export guidance

Skills

Build focused, versioned instruction capabilities:

  • Decide when to use a Skill, a main prompt, a tool, or a separate Assistant
  • Create, test, reuse, and restore Skill versions
  • Understand permissions, organisation publication, Spaces, Personal Chat, adoption, and audit history

AI Provider Configuration

Connect Eneo to AI providers and configure them per tenant:

  • Add providers and models through the Admin → Models wizard (OpenAI, Anthropic, Azure OpenAI, Google Gemini, Mistral and other LiteLLM-supported providers)
  • Self-hosted models via any OpenAI-compatible endpoint (for example vLLM)
  • Credential management, encryption, and shared vs. strict mode
  • Cost management and budget alerts

MCP Servers

Connect assistants to external tools while keeping discovery and execution under admin control:

  • Configure Streamable HTTP servers and authentication
  • Approve tool definitions before they reach the model
  • Forward user identity for user-specific catalogs
  • Preserve stateful MCP sessions across conversation turns

Web Search & Image Generation

Turn on the functions (Funktioner) assistants can use, and understand how they run:

  • Activate a search engine or an image source under Admin → Tools → Functions
  • Use a catalog image model directly, or an external MCP image server
  • Scope sources by user group, role permission and security classification
  • Edit and vary images with reference images, including follow-up edits

Deploy Eneo

Deploy Eneo in production environments:

  • Docker Compose reference stack with Traefik and Let’s Encrypt
  • Required secrets and environment configuration
  • Network isolation between proxy, data, and module tiers
  • Using nginx instead of Traefik
  • Object content storage options

Upgrading to 2.2.0

Prepare a deployment and its integrations for the names 2.2 no longer reads:

  • Environment variables to rename, and the ones removed without a replacement
  • What db-init, the backend and the web app log or refuse at startup
  • API names that changed for integrations
  • Commands, log names and open browser tabs after the upgrade

Upgrading to 1.7.0

Run the tenant-specific AI models migration safely:

  • What changes and which tables are affected
  • Backups, dependency chain, and ENCRYPTION_KEY prerequisites
  • Upgrade procedure, impact on running jobs, and credential handling
  • Verification, rollback, and troubleshooting

Upgrade File & Icon Storage

Run the one-time legacy content adoption safely:

  • Follow the offline db-init and online worker phases
  • Estimate PostgreSQL disk, WAL headroom, and wall time
  • Pause or restore when capacity runs low
  • Record a release-candidate result and verify completion before cleanup

Choose Content Storage

Choose and operate Eneo’s durable byte backend:

  • Use bounded PostgreSQL-inline storage by default
  • Run the bundled SeaweedFS service
  • Connect MinIO or another S3-compatible endpoint
  • Verify image provenance and readiness
  • Plan capacity, backup, restore, and troubleshooting

Document Processing

Upload and process documents for AI-powered knowledge bases:

  • Supported file formats (PDF, Word, Excel, PowerPoint) and size limits
  • Web crawling configuration and authenticated sites
  • How extraction, chunking, embedding, and retrieval work
  • Optimizing retrieval with chunk size and embedding model choices

SharePoint Integration

Connect Eneo to Microsoft SharePoint and OneDrive for document import:

  • Backend configuration and webhook secret
  • Microsoft Entra ID application registration and permissions
  • Service account vs. tenant app authentication
  • Configuring and verifying the integration in Eneo

Need more help?

If you can’t find what you’re looking for: