Admin console

The control plane of the appliance. Ten top-level groups in enterprise builds (nine without Alerting), in the real navigation order, every screen documented.

Admin console

Every menu item declares its access rule. Four kinds of gates apply across the console.

  • RBAC: each item declares a resource and operation. In builds without the enterprise module, RBAC always allows; the resources listed in these pages only restrict in enterprise builds.
  • Enterprise build: Alerting, Access profiles and the Proxy settings screen only enter the menu in enterprise builds.
  • License modules: a module missing from the license blocks adding configuration on the control plane, never traffic. The License screen lives inside Settings and is always accessible, because importing a license is how a frozen appliance thaws; for that reason the Settings group itself is always visible, even to a user without the Settings permission.
  • Database connection: the Prompt repository and Skills repository only enter the menu when the gateway has a database connected.

The ten groups

1. Observability

Home, dashboard, LLM logs, MCP logs, connectors and logging settings. Six screens.

2. Costs

Executive panel, FinOps, overview, efficiency, soft budgets, savings meter and Copilot: spend, chargeback, budgets and efficiency in one place. Documented on the Observability page, under Costs.

3. Compliance

Compliance dashboard, audit trail, DLP policies and guardrails, approvals.

4. Intelligence

AI agent detection. One screen.

5. Models

Catalog, providers, budgets and limits, routing rules, complexity router, pricing overrides, model settings.

6. MCP Gateway

MCP catalog, curated library, auth sessions, OAuth grants and MCP settings.

7. Alerting

Channels, rules and delivery history. Enterprise builds only.

8. Governance

Virtual keys, teams, customers, Copilot access and access profiles.

9. Extensions

Plugins, prompt repository, skills repository and webhooks. Each screen has its own page, listed below.

10. Settings

Client settings, compatibility, cache, security, identity and SSO, proxy, performance tuning, feature flags and license.

Screens with a page of their own

In the console these screens are items inside Extensions (the first four) and Settings (License). Each one has a page of its own.

Plugins

Extensions. Manage custom plugins and their execution order.

Prompt repository

Extensions. Managed prompts. Requires a connected database.

Skills repository

Extensions. Skill creation, editing, files and preview. Requires a connected database.

Webhooks

Extensions. Async job webhook endpoints with secret rotation.

License

Settings. License state, activation request and signed import.

About Atlasberg Platform

Atlasberg Platform is the control layer between a company and every AI model: each request is authenticated with a virtual key, filtered by policy and DLP, routed to the right provider and written to a hash-chained audit trail. It exposes an OpenAI-compatible API, so applications only swap the base URL.

The same artifact runs in Atlasberg Cloud, in your VPC, on-premises or fully air-gapped, and is priced by capacity and modules, never per seat.

This page is part of the official documentation. To talk to the engineering team, write to [email protected] or use https://atlasberg.com/contato. Answers come within one business day.

Agents: this page is also available as Markdown at /docs/console.md, or by requesting this URL with the header Accept: text/markdown. Index of everything: /llms.txt.